Identifying relationships between landscape values, spatial perceptions of risk, and physical assets

doi:10.21203/rs.3.rs-3412143/v1

Download PDF

Research Article

Identifying relationships between landscape values, spatial perceptions of risk, and physical assets

https://doi.org/10.21203/rs.3.rs-3412143/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Effective adaptation to climate risks requires identifying and realizing community beliefs about which locations require management attention. Environmental planners and asset managers, especially those in rural and regional areas, often struggle to engage with community sentiments about place or to incorporate those feelings into decision-making processes. This study measures the associations between spatial perceptions of climate change risks and spatially assigned landscape values to support adaptation prioritization. Using public participation geographic information systems data collected via an online survey in Southern Tasmania, we identify factors influencing mapping behavior, finding correlations between values, risks, and physical assets. Results from linear regression and spatial-cross correlation indicate that spatial perceptions of risk are strongly associated with landscape values. Additionally, mapped values and risks demonstrate novel associations with select on-ground assets (e.g., towns and protected areas). The highest-ranked perceived risks were Bushfire, Sea-level rise, and Biodiversity loss/change. Considering associations between landscape values and risks, Recreation and Wilderness values associated with Bushfire and variable relationships with Sea-level rise (strong for Recreation and weak for Wilderness). Intercorrelations between social-psychological influencing factors, values, and risks returned inconclusive results warranting further research. The approach described in this research has the potential to inform climate adaptation planning by categorizing key locations that require management action, potential areas likely leading to community resistance, and the underlying sentiments that inform landscape values and spatial perceptions of risk more broadly.

participatory mapping

landscape value

climate change

risk perception

geospatial analysis

PPGIS

Tasmania

Australia

rural geography

Climate change is increasing in intensity, frequency, and complexity (IPCC, 2022). Innovative community engagement methods are needed to identify the intersections between perceptions of climate change risks, landscape values, and potential policy responses to climate change. Environmental planners and asset managers often navigate competing needs of managing climate risks and responding to community concerns. Climate impacts and their high variability over space and time can disrupt how people think, feel, and act in relationship to the natural and social environments (Boulton, 2016). When residents do not see their views reflected in a plan, they can disengage from or actively resist management responses. While the role of place and landscape connection continues to receive attention (Hess et al., 2008; Lyth et al., 2016), how to accurately capture and reflect the places that people value, the threats they perceive, and potential trade-offs they may be prepared to make is challenging.

Even the best climate adaptation plans can be difficult to implement where community support is lacking, especially if people feel their views are not acknowledged or reflected in adaptation measures. Community connections, or the emotional bonds that people hold with places (e.g., sense of place, place attachment), are strong predictors of local conservation behaviors and can improve stakeholder buy-in and ownership of decisions about natural resource management (Blennow et al., 2019; Hess et al., 2008; Seymour et al., 2010; Williams and Vaske, 2003). Landscape values, for instance, are the features, both seen and unseen, that people assign and ascribe attachment and connection to (Brandenburg and Carroll, 1995). Landscape values provide a means to understand the spatial distribution of the types of values and attachments people hold and, therefore, might expect government action to protect in the face of threats, including climate change.

An essential first step in responding to climate impacts is understanding their spatial distribution to better target adaptation responses to physical assets and an awareness of community sentiment. While governments typically seek to incorporate the best available science, they are often hindered by limited resources and uncertainty about where to take action (Braunschweiger and Ingold, 2023). Governments must also grapple with climate change being simultaneously conceptually complex and inherently spatial (Morton, 2013). For example, the spatial distribution of risk for adaptation measures relies on climate projections and assessing potential on-ground impacts (Preston et al., 2011). However, this empirical distribution of climate risks in the landscape may differ from community perceptions of which places are at risk and the nature or source of those risks (e.g., the conceptualization of climate risk). Disconnects between community perceptions of risk and objective risks can negatively influence community support, undermining adaptation planning (Carmen et al., 2022).

This paper aims to identify the relationship between landscape values and spatial perceptions of climate change risk to inform climate planning decision-making better. To this end, we seek to answer the following questions: (1) Do social-psychological measurements of risk perception influence landscape values or spatial risk perceptions? (2) Are there statistical correlations between landscape values and spatial risk perceptions in places facing growing threats? (3) What are the spatial associations between all mapped values and risks and specific value and risk combinations? In answering these questions, we present a deeper understanding of the interconnections between the factors underpinning perceptions of climate change risk and their influence on landscape values. We also validate the mapped data by assessing the relations with physical assets. These findings can improve adaptation policies, reduce community resistance to climate action, and generate a social license to operate, especially in regional areas where financial and human resource constraints create the need to prioritize areas for action.

Landscape values and spatial risk perceptions are unevenly spatially distributed, and different types of people hold different values. Given this reality, there is a need to understand how values might correlate with risks from climate impacts and how these impacts might unequally impact certain socio-demographic groups. Raymond and Brown’s (2011) pioneering research laid the foundation for measuring the connections between landscape values and climate change risk by developing a novel methodology to collect, analyze, and communicate participatory mapping data. They focused on the relationship between the degree of attachment and perception of risk as a means for informing policy outcomes. Specifically, they concluded that the locations with high levels of value and perceptions of risk deserved the most resources and policy attention. Additionally, they looked at areas where values and risks overlapped with the land use type in those areas. While their research points to the broader need to reflect people’s perception of risk and the places they value in planning schemes, it assumed that all climate change risk perceptions were equal or could at least be weighed against landscape values for decision-making purposes.

Measuring landscape values and spatial perceptions of risk has predominantly been accomplished using public participation geographic information systems (PPGIS) or landscape values mapping, as demonstrated over numerous studies (Brown and Fagerholm, 2015; Brown and Kyttä, 2014; Brown and Weber, 2012). This methodology has included collecting mapping data through mail, interviews, group sessions, online, or a combination (Raymond and Brown, 2011). With the rise of digital, online, and internet-based GIS platforms (Babelon et al., 2017), collecting public participation data on landscape values and other socio-cultural features has become more accessible, even for small and under-resourced locations. The widespread use of GIS tools has led to the growing ability to map landscape values to interrogate what drives their spatial distribution and how these values overlap with risks, such as land use conflict and climate change. Despite the extensive use of PPGIS methods, limited examples specifically investigate the interaction of values and risks in dynamic rural regions. Additionally, there is a lack of testing value and risk perceptions against physical assets as a method of ground truthing to determine what place attributes drive community preferences and concerns and to further guide decision-making and prioritization of adaptation measures.

Community risk perceptions, recognized as subjective experiences that do not neatly map against the real-world experience of risk (Pidgeon et al., 1992), can be both barriers and opportunities for adaptation policy support (Etkin and Ho, 2007; Singh et al., 2017). Risk perception frameworks are informed by various cognitive factors, experiential processing, socio-cultural influences, and socio-demographics as the main drivers of climate change risk perception (van der Linden, 2015). These factors underpin the multidimensionality of risk, whereby an individual’s perception of climate change risk varies both from the perceived spatial manifestation of risk and the scientific projections of risk. This risk perception gap within coupled socio-ecological systems is informed by and informs assigned landscape values, such that the measurement of one without the other will further contribute to the uncertainty issues hindering climate action. By measuring landscape values, spatial perceptions of risk, and perceived (subjective) climate risks in tandem, this paper provides a clearer understanding of places most and least likely to evoke community resistance and support to climate action.

To answer our research questions, we implement a variation of Brown's (1984) participatory mapping procedures that measure both assigned landscape values (e.g., assigned worth given to places by people) and spatial risk perceptions (e.g., a grounded expectation of climate impacts). We conducted our work in the Huon Valley, Tasmania (see Supplementary Material for step-by-step methods).

3.1 Study location – Huon Valley coastalscape

The Huon Valley coastalscape is a dynamic rural region comprised of diverse coastal and terrestrial features, a distinct yet shifting community composition, and complex climatic risks threatening industries, individuals, and ecosystems, thus making it an ideal case study area. The Huon Valley coastalscape is the southernmost region of Tasmania (Fig. 1a), comprised of the lands and waters of the Huon Valley council area (Fig. 1b). The municipal land area covers 5,497 square kilometers. It is home to more than 16,000 residents living in the five main townships (Huonville, Franklin, Cygnet, Geeveston, and Dover) along the Huon River and D’Entrecasteaux Channel (Fig. 1c). Most of the region is characterized by the vast native forests of the Southwest National Park, which is part of the Tasmanian Wilderness World Heritage Area. The remaining land coverage includes extensive forestry, pastures, orchards, vineyards, and the urban and other “intensive use” classified areas. The aquatic assets of the region include numerous wetlands, salt marshes, lakes, ponds, rivers, rivulets, inland caves, and the estuarine and marine waters that interface with the coastline to distinguish the region.

Similarly, the Huon Valley is defined by a relatively diverse set of cultural backgrounds and leading industries. Most of the region's residents are older and have longer connections to the landscape. However, in the last several years, it is anticipated that younger individuals are moving into the region either to have a plot of land or to escape the city issues like rising rent and increased density, both in Hobart and from the mainland (Denny and Pisanu, 2019; Osbaldiston, 2022). The region also has a considerable diversity of education levels, including a largely illiterate population and those with post-graduate degrees. The leading industries in the region are agriculture, forestry, and aquaculture, all of which rely heavily on the natural resources that define the coastalscape. While there seems to be relatively high support for the agricultural industries (e.g., orchards, vineyards, cattle, sheep), the other primary industries continue to be controversial in the eyes of Huon Valley residents.

Climate change projections for southern Tasmania have been extensively researched compared to its size (Remenyi et al., 2020). The main predictions include drier summer and wetter winter months, contributing to an increased risk of bushfires, floods, and droughts. Additionally, sea level rise and increased storm surge will act as a risk multiplier. However, low-lying areas, mainly heavily developed, are relatively low risk compared to other parts of Tasmania and abroad. While these projections may seem dire, the increased climate risk across the island is significantly less than in most parts of the world, leading some researchers and journalists to suggest the state is a “climate lifeboat” (King and Jones, 2021). Climate adaptation in the Huon Valley is somewhat contested, and at the time of writing, the region lacks a comprehensive climate change adaptation plan.

3.2 Sampling technique – online mapping platform

The data collection phase for this study took place during the height of the COVID-19 pandemic and local lockdowns, so we opted for a purely internet-based approach to avoid the challenges of in-person research (Peyton et al., 2022; Singh and Sagar, 2021). Following a review of existing online participatory mapping options (Cochrane and Corbett, 2020), we determined that developing our platform would be more cost-efficient and could be used for multiple studies. The platform functioned like those utilized by past studies (Brown and Kyttä, 2014; Fagerholm et al., 2021), wherein participants would respond to survey questions and drop pins on a map based on landscape values and climate risks.

Recruitment for the survey took place over two months following the Dillman method (Dillman et al., 2014), utilizing various communication methods, advertisements, and announcements. The first round of communication came from the Huon Valley Council, which advertised the survey in emails and on its website. Community groups were contacted early on and asked to share an advertisement for the survey via social media and other platforms rather than paying for an advertisement campaign. Repeated social media posts occurred throughout the recruitment period. Also, poster advertisements were placed on notice boards across the region in the first week, specifically outside community halls. At the end of the first month, mailers were distributed to select regional postcodes, an advertisement was taken out in the local newspaper, and a short ad was presented on the local community radio station. The response rate was monitored regularly, and response quality was checked to ensure robust data, gauge advertisements' efficacy, and limit the potential for fraudulent responses (Author(s), XXXX).

The online mapping survey included three sections. In the first section, participants answered a series of socio-demographic questions and several questions related to their connection to the region. Rather than excluding participants based on the postcode of their home address, we sought respondents with any connection to the region (e.g., living, working, spending time in, etc.), thus avoiding concerns around land ownership as the primary means of landscape connection. In the second part, participants placed at least five landscape value points at a 5km scale, meaning they had to zoom in to a particular scale before placing their pins to ensure consistency in data analysis. Then, on a second map, participants were asked to place any number of “places at risk” points. In both cases, participants were encouraged to provide text for each point but not required, yet most respondents included text on 10–100% of their pins.

The final part of the survey had participants answer questions from three separate Likert scales. The first scale (questions 13–24) was a multi-part scale that combined both place identity and place dependence questions based on previously established in the literature (Brown and Raymond, 2007; Raymond et al., 2010). The second scale (questions 25–38) asked about climate change risk perceptions, including questions from the NEP scale and risk Perception Index (Dunlap, 2008; van der Linden, 2015). This scale included questions about the main influences of risk perception outlined in van der Linden (2014), though the limited number of questions due to an already expansive survey. The third scale (questions 39–45) explored the issue of future climate impacts by presenting a projected climactic change and asking participants to identify dissatisfaction with the likely change.

3.3 Analysis techniques

After assessing respondents for validity, the first author entered the survey data into SPSS® and digitized the landscape value and climate change risk points into ESRI ArcGIS® software. Spatial data and survey responses were combined and used in descriptive and inferential statistics across the data set.

3.3.1 Respondent profiles and connections to place

Comparing respondent characteristics from the survey sample to the Australian Bureau of Statistics regional census data (Australian Bureau of Statistics, 2021) helps to determine the sample's representativeness. A combination of t-tests and ANOVA in SPSS® explored the differences in the mean number of mapped landscape values and climate risks by survey participants across various socio-demographic characteristics. Participants answered questions designed to determine their connection to the region, including the length of their connection in years and their perceived knowledge of the region. We also asked about their experiences with extreme weather events and the perceived likelihood of future extreme weather events. Both these features were tested for potential correlations with other variables using a combination of descriptive statistics (mean and standard deviation) and intercorrelations (Pearson’s) in SPSS®.

3.3.2 Attachment to place, risk perception, and scale interconnection

To analyze the various social, experiential, and psychological measurements from the survey, a series of scales were created and analyzed through a combination of descriptive statistics (mean and standard deviation), intercorrelations (Pearson’s), and scale reliability (Cronbach’s alpha) in SPSS®. ‘Time lived in’ and ‘knowledge of’ the region were 5-point Likert-scale questions. Experience with extreme weather events was asked as a multi-choice question and coded as a three-point scale (0 = no experience, 1 = one experience, and 2 = more than one experience). The likelihood of experiencing future extreme weather events was also on a 5-point scale ranging from extremely unlikely to extremely likely. The perception of political effort to address climate change was a two-question scale that included Likert-scale questions about state and local council efforts.

The remaining three scales analyzed include the ‘dissatisfaction with possible climate changes,’ the combined ‘place identity’ and ‘place dependence,’ and the altered ‘risk perception’ scales. The first scale included seven questions following the statement, “These are likely scenarios determined by Climate Futures.” Each question presented a scenario with five options ranging from “very satisfied” to “so dissatisfied I would leave the region.” The two place attachment scales included six questions, shortened from the standard place attachment scales (Brown et al., 2015). Each scale was analyzed, and the questions were combined to determine reliability and correlations. Lastly, the risk perception index included twelve 5-point Likert-scale questions with options for each based on the specifics within the questions.

3.3.3 Statistical relationships between values and risks

We calculated the overall rank based on total markers placed and mean importance to identify the relative importance of each landscape value and climate risk to respondents. Mean importance was then calculated by dividing the total number of markers placed by an individual by the total number of markers placed for that category and then taking an average across the sample. A fractional approach determined the overall rank, with each value and risk type ranked ordinally from highest to lowest number of points placed across all respondents, ranking each mean importance via a similar scale and then adding these two scales together. Since participants could mark any number of responses, it was necessary to calculate this overall rank versus ranges, totals, or variances.

To examine the general distribution of landscape values and climate risk hypotheses, we used nearest neighbor analysis to determine if specific sets of points locations were completely spatially random (CSR) on the landscape. Rejection of the null hypotheses would indicate that the points are not random, whereas acceptance may indicate some degree of randomness. The nearest-neighbor statistic (R) measures the points' spatial distribution calculated by dividing the average Euclidian distance of all points within a specific polygon by the expected distance of points under an assumption of random distribution. R values closer to 0 indicate more clustered distributions, whereas more randomly distributed points will return an R-value closer to 1. Since some sample sizes for individual risk categories were relatively small, we expected the degree of randomness to be higher than in similar studies elsewhere.

Using multiple regression analysis in SPSS®, we determined the relationships between landscape values and climate risks. The first regression analysis assessed the latent, non-spatial association between landscape values and climate risk mapping based on the number of markers placed for each category by individual respondents. The overall importance rank of each landscape value served as independent predictor variables in the regression, with the overall rank of climate risk as the dependent variables. The analysis provided the relative relationship between values and risks (R and R²) and the predictor significance (β) for each value and risk. We tested the total number of markers placed instead of the overall importance rank as the independent variable(s), as is standard for PPGIS research (Raymond and Brown, 2011).

Raster analyses determined the spatial relationship between landscape value and climate risk densities. Specifically, spatial cross-correlation analysis calculated Spearman’s Rho correlation coefficient (r) of the overlap of two raster densities at randomly selected points. Using the kernel densities developed for each landscape value and climate risk, we extracted and associated the density values for points to calculate coefficients between landscape values and climate change risks. Cross-correlation results indicate the level (with higher r values indicating stronger correlations) and direction of the spatial relationship between value and risk densities.

3.3.4 Spatial distribution of landscape values and climate change risks

To assess the spatial distribution and test assumptions that points would be associated with on-ground features (e.g., protected areas, towns, popular recreation locations, etc.), maps were generated in ArcGIS®. Using 500 m grid cell sizes with no search radius and square map units based on Planar densities, kernel density maps were created. This grid cell size and search radius combination is consistent with existing literature, provides a reasonable threshold for measuring density, and is appropriate within the study’s scale. A Tasmanian-wide boundary ensured that outputs were consistent across analyses. These densities were calculated for both the entire value and entire risk data sets, a combined total values and risks set, and for each value and risk type. The values and risks were classified using the Jenks natural breaks classification method, which breaks the densities into classes, reducing the variance within classes and maximizing the variance between classes. We then aggregated the classes and rasterized them into high, medium, and low grids. We then used the raster calculator function to generate spatial intersections of low, medium, and high values and risks, with the resulting raster layer including nine classifications. The produced maps were visually assessed for spatial associations based on critical study factors and known on-ground assets identified through their location in the region, mentions during mapping procedures, researcher familiarity with the landscape, and combined spatial intersection values.

The following section presents results from statistical analyses across the study variables. See Supplemental Material for the comprehensive set of statistical outputs beyond the following key results.

4.1 Respondents

After cleaning all data and retaining only complete quality responses, the total number of respondents was comparatively smaller than in other studies, limiting some statistical analyses of the data; however, the total number of points was sufficient for established PPGIS approaches (Brown, 2017). A total of 39 individuals fully completed the survey (i.e., mapped the minimum number of points). Of those individuals, a few (n = 5) finished parts 1 and 2 of the survey and skipped over the remaining questions before submitting. One thousand four hundred forty-three digitized points make up the spatial data set, including 1,046 landscape value points and 397 places at risk points (see Supplementary Material for full socio-demographic results).

Most respondents were over 55 (61.6%) and female (64.1%). They were also mostly university graduates (64.1%) with incomes over $25K (59%). Based on census data, there was a very high overrepresentation of high-income earners and university-educated respondents. Older respondents and females were also relatively overrepresented, though this is expected based on the people most likely to respond to surveys. The most underrepresented groups include people under 24, those identifying as Aboriginal or Torres Strait Islander, and those without university degrees or TAFE certification. However, those groups are likelier to have low literacy levels and a relatively high proportion of Huon Valley residents.

Almost all the respondents (95%) included permanent residence as a connection, with the remaining two indicating “have close family in the region” and “spend time doing recreation there” as their only connection. Nine respondents also included their main livelihood, business ownership, family, and an “other connection” in addition to their permanent residence connection. Most respondents (59.4%) lived in the region for more than ten years, including a few (n = 3) who lived there their whole life. The distribution of respondents’ knowledge of the region was relatively normal, with most (84.5%) indicating at least a “good” level of regional knowledge.

Most respondents (84.6%) responded that they had experienced at least one type of extreme weather event, with the vast majority indicating that they had experienced more than one type of extreme weather event or multiple of the same type (i.e., multiple bushfire events). Almost all of the participants (82%) responded that they believed it was “likely” or “extremely likely” that they would experience extreme weather events in the future, with only one respondent suggesting the future experience to be “extremely unlikely.” Six respondents (15.4%) did not answer these questions related to extreme weather events.

4.2 Social, experiential, and psychological factors

4.2.1 Scale reliability and correlations

Each question set was tested in SPSS® to determine the scale reliability and associations between factors (see Supplementary Material for descriptive statistics and results from scale-based linear regressions). Cronbach’s alpha scores closer to 1 indicate the high reliability of the scales. All scale variables returned alphas greater than 0.8, which indicates a high reliability of the scales, including risk perception (alpha = 0.98), place attachment (alpha = 0.911), and future dissatisfaction (alpha = 0.859).

Most predictor variables were positively correlated, excluding political perceptions, which were negative for the time in the region and all the scale variables. In contrast, only a few variables were statistically significantly correlated. Of the experiential variables, ‘knowledge of the region’ was correlated with ‘time lived in the region’ (r = 0.45), both the combined and two separate place attachment scales (r = 0.44) and the risk perception scale (r = 0.36). The perceived likelihood of experiencing future extreme weather correlated with multiple variables, including past experiences (r = 0.41), perception of political effort (r = 0.37), and dissatisfaction with potential future impacts (r = 0.52). Future experiences are also correlated with the risk perception scale (r = 0.73); however, this is expected as the risk perception scale includes similar questions about future extreme events. The dissatisfaction and risk perception scales also showed a strong correlation (r = 0.52), as did all three place scale variables.

4.2.2 Associations between scales, values, and risks

Results for all three scales in regression models were not statistically significant, with landscape values explaining a low to moderate amount of the variance for place attachment (R² = 0.314), risk perception (R² = 0.326), and future dissatisfaction (R² = 0.267). None of the landscape values appeared as significant predictors of any of the three scales. We also ran the analyses based on the number of mapped values, as is standard in the literature. However, these returned a similar lack of statistical significance. The collinearity diagnostics on the regression models suggest a tolerable level of multicollinearity in the independent variables with VIF values below the general threshold of 10 for obvious concern (Myers et al., 2000).

Regression model results for the scales were also not statistically significant for climate change risk. However, climate risks explain a low amount of variance for place attachment (R² = 0.280) and a moderate amount for both risk perception (R² = 0.400) and future dissatisfaction (R² = 0.449). Biodiversity change and Bushfire emerged as significant predictors for place attachment and risk perception scales, respectively. The analyses were based on the number of mapped values, as is standard in the literature. However, these returned a similar lack of statistical significance.

4.3 Correlations, associations, and relationships between values and risks

4.3.1 Relative importance of values and risks

The total number of points, percentage of all landscape value points, and percentage of all mapped points for the top landscape values are Recreation (n = 137; 13%; 9%), Aesthetic (n = 132; 12%; 9%), and both Wilderness and Economic (n = 117; 11%; 8%). For climate change risk points, the top three are Bushfire (n = 181; 45%; 12%), Biodiversity loss (n = 60; 15%; 4%), and Sea-level rise (n = 36; 9%; 2%). The remaining eight landscape values and six climate risks all represented less than 10% of the values or risk, respectively, as well as less than 7% of the total digitized points for values and less than 5% of the total risks. The overall highest set of points was the climate risk Bushfire; however, this was partially due to a single respondent mapping 65 Bushfire points in the session.

The overall importance ranks are presented below to account for differences in mapping effort and the total number of points between the two categories (Table 1). While Recreation was the most frequently mapped, Wilderness was calculated as having a higher mean importance for individuals, whereas Aesthetic had a lower mean importance. Bushfire was the most frequently marked and held the highest mean importance, even when adjusting for the abovementioned outlier.

Table 1

Importance rank of landscape values and places at risk

n ^a

Mean importance ^b

Overall Rank ^c

Landscape values

Recreation

137

5.44

Wilderness

117

5.56

Economic

117

4.01

Aesthetic

132

3.90

Social

3.38

Intrinsic / Existence

2.04

Heritage / Culture

2.87

Biological

2.01

Life-sustaining

1.55

Learning / Research

1.71

Spiritual

1.52

Special ^d

n/a

Places at risk

Bushfire

181

9.85

Biodiversity loss

6.48

Biodiversity change

4.41

Sea-level rise

3.00

River flooding

4.13

Land erosion

1.96

Coastal erosion

1.44

Storm surge

1.41

Other risks

1.31

^a Total number of markers placed

^b The mean respondent importance score is based on the fractional ranking of the total number of points placed for each type by an individual.

^c Overall rank is based on a combination of total markers (n) and the mean importance

^d Special places were not included in the importance ranking because all respondents were required to place at least one special place

4.3.2 Average nearest-neighbor

The average nearest neighbor tool in ArcGIS® was used to test the null hypothesis of Complete Spatial Randomness (CSR) of point distribution (see Supplementary Material). The nearest neighbor calculates the spatial distribution of points with more randomly distributed points returning an R-value closer to 1. Two landscape values (Biological and Life-sustaining) and four climate risks (Biodiversity change, Coastal erosion, Land erosion, and Storm surge) returned high R values (R > 0.9), indicating high degrees of randomness and rejection of the null hypothesis. We suspect this relates to the number of markers dropped for each category and how participants interpreted each category. The text accompanying those marker types informed the latter and the former by reviewing regional marker placements. Social, Heritage/culture, Economic, and Learning/research landscape values had the lowest R values (R < 0.6) and, therefore, the least random clustering. This expected lack of randomness is because these values tended to cluster exclusively around the towns. The least random climate risks were River flooding, Sea-level rise, and Bushfires.

4.3.3 Statistical correlations

The following regression model identifies the non-spatial relationship between landscape values and climate risks (see Supplementary Material for complete statistics). The number of mapped landscape values were moderate predictors for the number of mapped climate change risks, regardless of spatial location (R² ranges from 0.28 to 0.66). For example, individuals who map more Aesthetic, Biological, Economic, and Heritage/culture values tended to map more Biodiversity change risks. The number of mapped Aesthetic values was also a significant predictor of the number of mapped and Sea-level rise (β = −0.93); Social values were significant for Bushfire (β = 0.72) and Land erosion risks (β = 0.38).

The spatial relationships between landscape values and climate change risks were examined using a raster spatial model. Spatial cross-correlation analysis was performed on density grids generated for landscape values and climate risks, with all results significant at p < 0.01 unless otherwise indicated (Table 2). There are very strong spatial associations between Aesthetic and Sea-level rise (r = 0.75), Biological with both Biodiversity change (r = 0.77) and loss (r = 0.76), and Heritage and Sea-level rise (r = 70). Recreation had strong correlations with Biodiversity change (r = 0.71), Bushfire (r = 0.73), and Sea-level rise (r = 0.73). Recreation was among the top-ranked spatially correlated landscape values across all risk categories, excluding river flooding. Despite being the second highest ranked value, Wilderness had comparatively low spatial correlations, likely due to the distribution of those values across the region compared to concentrations of other values around the river and townships.

4.3.4 Spatial association between select values and risks

To further illustrate the spatial associations between landscape values and spatial perceptions of climate change risk, we developed a series of density maps based on combinations of values and risks informed by the preceding analysis. We demonstrate two risks for several reasons, including typically being identified as the most prominent threats to the region, Bushfire as the top-ranked risk, and Sea-level rise for its importance in local planning and adaptation. The two other top-ranked risks (Biodiversity loss and Biodiversity change) were excluded due to a combination of spatial randomness and limited local government ability to manage. When then compared the risks with two values, based on the overall rank, significant linear regressions, and strength of spatial cross-correlation. A third value, Wilderness, was also included in the spatial association mapping due to its overall importance and stated relevancy across community sentiment and council descriptions. All the values demonstrate relationships with and beyond the two risks and are described by community members as some of the main drivers for their attachment to the region.

Bushfire risk (Fig. 2a) has a strong non-spatial relationship with Social, a strong spatial relationship with Recreation, and a medium spatial relationship with Wilderness. Social values (Fig. 2b) are concentrated around the region's major towns (Huonville, Franklin, Geeveston, Cygnet, Dover, and Ida Bay). Apart from Cygnet, all high Social value areas were within high Bushfire risk areas (Fig. 2c). For Recreation (Fig. 2d), Bushfire is perceived as posing the highest risk to areas from Port Huon to Huonville and in dry eucalypt forest areas near Surveyor’s Bay (Fig. 2e). Areas perceived as having medium risk include Hartz Mountain, Pelverata, and select areas along the Huon River. Wilderness values (Fig. 2f) shared high-value/risk areas exclusively with Hartz and Tahune. Medium-risk areas include Franklin Forest, Lonnavale, and the area around Pelverata and Snug Tiers (Fig. 2g). While bushfires are typically described as a significant risk to the wilderness areas of the region (Grose et al., 2010), the spatial perception of bushfire risk seemed to anchor more around communities and towns.

Sea-level rise (Fig. 3a) has a strong spatial and non-spatial relationship with Aesthetic values, a strong spatial relationship with Recreation, and a limited spatial and non-spatial relationship with Wilderness. Sea-level rise risk was primarily assigned to the Huon Estuary conservation area, the Huon River near Surges Bay, and Southport. Aesthetic values (Fig. 3b) are concentrated around the coastal areas along the Huon River and in select locations across the region (e.g., Hartz Mountain). Sea-level rise was seen as the biggest threat to the Franklin, Egg Island Conservation area, and Huonville areas, as well as specific beaches (Fig. 3c). Similar locations were seen with Recreation (Fig. 3d), where the Huon River areas saw high Sea-level rise risk (Fig. 3e), likely based on the high incidence of boating activities identified during the mapping process. Lastly, for Wilderness (Fig. 3f), there was limited overlap with Sea-level rise (Fig. 3g). Wilderness value more broadly was not assigned to coastal or riparian areas except for Ida Bay and the Southport Lagoon.

4.3.5 Spatial association of all mapped values and risks

After generating kernel densities for all landscape values (Fig. 4a) and places at risk (Fig. 4b), we developed a combined density map that illustrates the spatial associations between areas with various combinations of high and low landscape values and places at risk (Fig. 4c). We then determined the key on-ground assets for each location with some degree of noticeable density. In our case, using a 3-point scale for each density map and a 9-point box grid for the overlapping areas, any location with a combined medium result was checked for potential associations. The main assets included towns and protected areas with other assets, including waterways (e.g., lakes, creeks, waterfalls), green spaces (e.g., forests, bays), and recreational locations (e.g., walks, caves, historical sites).

On the southern edge of the landscape is the South Coast Track, one of the region's most popular multi-day tracks, along which many density points were placed that were not readily associated with other assets. Another location with a unique density pattern was the Southwest National Park. This density cluster was not clearly related to a distinguishable asset. However, when reviewing the mapping platform used for data collection, we found that the system labeled the national park in that specific location, likely drawing participants to place markers near the text rather than associated with something on-ground.

The highest density locations for both landscape values and places at risk were closely associated with the towns adjacent to the Huon River (i.e., Huonville, Franklin, Port Huon, and Dover), where respondents were located. Exceptions include Hartz National Park, the Tahune Airwalk, and surrounding protected areas. Of particular interest are the areas beyond the towns that were marked as being medium to high risk but with limited to no value markers. Similarly, areas around Recherche and the south coast had high levels of value but limited numbers of risk markers, indicating the need for further investigation and consideration about planning decision-making.

The primary aim of this study was to identify the spatial and non-spatial relationships between assigned landscape values and spatial perceptions of climate change risk in the Huon Valley coastalscape of Southern Tasmania to improve climate adaptation decision-making through reduced community conflict. In this pursuit, we answered three research questions, which we now address in turn. (1) Do social-psychological measurements of risk perception influence landscape values or spatial risk perceptions? In testing the influence of social-psychological correlates, precisely that of risk perception as defined by van der Linden (2015), there was limited evidence to indicate a strong relationship between the risk scales and mapping behavior. However, evidence suggests that landscape values are influenced by levels of risk perception, thus signaling the need to understand better the role social-psychological factors play in people’s conception of place and risk. (2) Are there statistical correlations between landscape values and spatial risk perceptions in the study area? Similar to other studies (Moser, 2014; Raymond and Brown, 2011), our analysis suggests a unique relationship between landscape values and spatial perceptions of risk in the case study region. In addition to the overall correlations between the landscape value typology and spatial risk categories, we found that relationships were particularly strong for specific value and risk combinations. These pairings sensibly align with the types of underlying features that would drive both the creation of value or attachment and exposure to risk – for example, recreational opportunities in wilderness areas at high risk of bushfires or values associated with recreating with family at foreshore areas at risk of sea level rise or flooding. (3) What are the spatial associations between all mapped values and risks and specific value and risk combinations? We determined that both the overall spatial associations and specific value/risk associations highlight key locations across the region that require both management priority and reframing of local council action and that the pairings provide both the spatial priorities as well as the types of actions needed to mitigate risks and manage values. While it is safe to assume that locations with combined high landscape value and high spatial risk perception will require increased management effort, their connections to specific values and risks offer both a challenge and an opportunity for councils to reframe their adaptation responses.

5.1 Influence of social-psychological correlates

Data analysis suggests that despite the high reliability of the social-psychological scales and their intercorrelations, their influence on assigned landscape value and spatial risk perception mapping behavior is not as strong or direct as hypothesized. Even the combined place attachment scale, which has previously demonstrated an influence on mapping behavior (Brown et al., 2015; Brown and Raymond, 2007; Williams and Vaske, 2003), yielded relatively few statistically significant outcomes in both spatial and non-spatial calculations. The other scales, specifically the adjusted ‘climate change risk perception scale’ and ‘future dissatisfaction scale,’ resulted in limited statistical significance. However, there were several notable correlations with mapping behavior. For example, an individual’s choice to map a particular number of climate change risk markers correlates with high risk perception levels and future dissatisfaction. Although for specific values and risks, the strength and direction were neither uniform nor predictable, thus indicating a complex relationship between the social-psychological correlates at the place-specific scale.

Our findings echo Raymond and Brown's (2011) recommendation of incorporating psychological factors better to understand drivers of risk for adaptation policy decisions. Researchers continue to demonstrate the complex relationships between perceptions of climate change risk and place attachment (Bernardo, 2013; Bonnes and Lee, 2017; De Dominicis et al., 2015; Etkin and Ho, 2007; Praskievicz, 2022), yet the divergences between different types of risk perception are not clearly reflected in the literature. For example, we attempted to measure and compare two types of risk perception: spatially assigned risk (e.g., specific locations perceived to be at risk due to climatic or other factors) and social-psychological determinants of risk or risk worldviews (e.g., internal thoughts and feelings about risk more broadly influenced by various social and nonsocial factors). While both seem to demonstrate a strong relationship with place attachment, as demonstrated in our research and other studies, their intersected influence is less easily described. Nevertheless, exploring the connections between the underlying influences on value and risk mapping provides a more meaningful understanding of why people place risks and further highlights the need to explore other social-psychological influences of mapping behavior.

5.2 Correlation between values and risk

Our data indicates that the decision for an individual to map specific values predicted the number and type of mapped risks in some cases. For example, an individual's likelihood to map several aesthetic values significantly predicted the number of mapped biodiversity changes, river flooding, and sea-level rise risk points. Across the data set, some values were good predictors of risk mapping behavior, whereas others were not significant. Individual and collective responses indicate that the importance and decision to map types of values and risks are related to factors beyond the type of mapped feature or the specific location being mapped.

Recreation was the most important value both in terms of total markers and mean importance but showed limited influence on the likelihood of an individual mapping risk. Furthermore, landscape values that demonstrated predictive capabilities for the number of mapped risk locations, such as Aesthetic or Biological, were not noticeably more likely to share high levels of spatial association with the paired risk types. These findings closely align with previous studies (Raymond and Brown, 2011). Further complicating these results is that the values and risks demonstrated a high degree of Complete Spatial Randomness while tending to show the strongest spatial and non-spatial relationships with other variables. While this may partially be a product of smaller sample sizes for marker types or the structure of the case study region, it also highlights the complex correlations between values and risks.

Across the different statistical analyses, the most relevant values (as determined by mean importance, spatial randomness, and spatial/non-spatial correlations) were ultimately Recreation, Aesthetic, and Wilderness. These results reflect how people describe their reasoning for relocating to the region (Denny and Pisanu, 2019; Osbaldiston, 2022). For risks, the most relevant were Bushfire, Sea-level rise, and Biodiversity loss. These risks are typically described as being the most prescient for Tasmanians (Lyth et al., 2016), with bushfires being the most likely risk to have previously experienced (Buys et al., 2012), while both sea-level rise and biodiversity loss are commonly discussed as risks in Tasmania (Grose et al., 2010). In most cases, these relevant values and risks had medium to strong intercorrelations, excluding Wilderness and Sea-level rise, where the former was marked as an exclusively terrestrial landscape feature. Linking climate adaptation actions for the three main risks to the relevant landscape value type aligns with values-based approaches widely seen as critical to increasing community acceptance of local climate change actions (Adger et al., 2013; Artelle et al., 2018; Persson et al., 2015). Individual and compound climate risks manifest in places and features distinctively and thus impact specific values in different ways (Hess et al., 2008). Our results indicate that people are uniquely suited for identifying links between specific risks and assets, primarily due to their knowledge about the places they value and which threats are likely to be present. Decision makers must consider the unique combinations of assigned values, perceived risks, and landscape assets, as each requires discrete adaptation actions.

5.3 Spatial associations of select risks and values

While there was a high degree of variability in the mapping behavior of participants, spatial patterns emerged in all analyses of landscape values and spatial risks. Values and risks are not randomly distributed across the region, including those demonstrating some degree of spatial randomness. In fact, they tend to cluster around on-ground assets such as towns, protected areas, or other essential features. For example, as shown in Fig. 4, areas with high value and high spatial risk perception are most prominently around the townships along the Huon River, Hartz Mountain, and areas around the Southwestern National Park. When it comes to decision-making around climate change at the local level, many of these areas would already be of high interest to planning authorities due to existing population concentrations or identification as critical tourist infrastructure, as would be the case for Hartz and Tahune. This is a significant area for concern as many of these locations have recently been determined to be most suitable for future development (Huon Valley Council, 2023), while limited adaptation planning remains in place. It suggests that the council needs to develop a more robust rationale for steering development into at-risk places or quarantining risky places from future development (Braunschweiger and Ingold, 2023).

The spatial distribution of both Recreation and Aesthetic values tends to be assigned to areas under the local council's authority. In contrast, Wilderness appears in areas beyond their authority to act, meaning that jurisdiction is under state or national authority. When implementing a values-led approach to adaptation (Artelle et al., 2018), local councils must focus on the values most aligned with their authority to act and the specific action type. For example, the Huonville Foreshore, a highly valued local area at risk of sea-level rise and flooding, is under local management compared to the Egg Island Conservation Area, an equally highly valued area with similar risk, which is under state control. Both elicit strong sentiments from community members for protection but receive differing levels and forms of adaptation from the council. In some cases, councils may need to pressure higher levels of government to ensure local landscape values are duly conserved (Wewerinke-Singh and Salili, 2020).

For areas with specific combinations of relevant high values and risks (including Bushfire and Sea-level rise), there are several variables and issues for planning authorities to consider, which relate directly to conceptualizations of climate change risk more broadly (Buys et al., 2012). For example, the council struggles with mitigation bias (Measham et al., 2011). Politically, council factions are driving mitigation responses (e.g., using electric vehicles) despite their limited contribution to national and global emissions. However, adaptation to select risks or key locations is lacking. Like many local governments and rural areas, the council will be forced to grapple with the decision to defend in place or undertake managed retreat (Abel et al., 2011), regardless of any other prior climate actions. The response to bushfire risk in recreation value areas will likely require a combination of bushfire preparedness (Frandsen et al., 2011), managing risk (Neale et al., 2019), and reframing the relationship between Bushfire, landscape, and community (Kruger and Beilin, 2014; Reid et al., 2020). The response to risks in wilderness areas typically falls under a combination of preservation and restoration (Mendel and Kirkpatrick, 2002; Wilson et al., 2007). However, as is the case in Huon Valley, these areas are often beyond the scope of the authority of local councils, or the efforts are mismatched with the threat level (Mendel and Kirkpatrick, 2002; Rands et al., 2010).

5.4 Implications for adaptation planning

We have identified several challenges for adaptation planning at local levels. We concur with Raymond and Brown (2011) that government agencies, particularly at the local level, have limited resources and thus must target priority areas with shared high value and high risk. By assessing and reflecting on specific landscape-level assets and unique combinations of values and risks in relation to those assets, our approach builds on their methodology to provide more precise and policy-relevant recommendations. As such, it would not be controversial to suggest that the Huon Valley Council targets the priority areas described in the preceding sections, especially since these locations seem to both fall under the council's purview and have established resource needs.

For example, our data indicates that people perceive their recreational values as most at risk from bushfires or sea-level rise. Recreational values tend to cluster around towns and community centers. While bushfires and sea-level rise overlap in select locations with recreational values, their presence is a significant concern across the region. For decision-making, without considering how risks and values anchor to specific physical assets, there is a possibility of expending resources to protect Recreational value areas or to reduce the risk of either bushfires or the consequences of sea-level rise (i.e., coastal inundation). Since the Recreational values tend to cluster with several other values, their perception of risk and loss to the community could be overestimated. Additionally, the consequences of landscape value loss are not proportional to the actual loss from climate change impacts, possibly forcing the need to prioritize specific value types over others.

Furthermore, there are significant differences between individual spatial perceptions of risk, the discursive conception of risk, and the projected on-ground risks, with the latter being the primary source of information for decision-making. This study's primary area of focus was the overlap between spatial risk perceptions and landscape values and limited attention to the effect of climate risk discourses on spatial perceptions of risk. People’s climate risk discourses are not exclusively based on a comprehensive understanding of climate change impacts, as indicated in the risk perception literature (Adger et al., 2013; McDonald et al., 2015; O’Neill and Nicholson-Cole, 2009; Rawluk et al., 2019; van der Linden, 2015). Since we cannot assume that spatial perceptions of risk reflect projected risk but are a byproduct of their risk discourses, and decision-makers rarely utilize community understandings of risk in policymaking, a gap must be addressed to reduce conflict and increase understanding of climate actions.

For instance, bushfires were the most widely placed climate change risk, even when accounting for outliers. Nevertheless, the risk markers tended to concentrate closer to values and along the Huon River, which, while certainly at risk of Bushfire, is significantly less so than bush areas that make up most of the region. The higher perception of bushfire risk may be more experientially related to recent bushfire events either in the region or across Australia, cognitively related through the widespread focus of Bushfire as a significant risk both presently and increasingly into the future throughout Australia, or socio-culturally related to people’s feelings around the potential impact to themselves and their more-than-human community (Booth and Harwood, 2016; Lange and Gillespie, 2023; Neale, 2018).

5.5 Future research opportunities

Both the extent of climate risk discourse influence on spatial risk perceptions and the gap between spatial risk perceptions and projected climate risk are topics suitable for future research. While identified as having limited evidence in this study of the role climate risk discourses play in mediating an individual’s mapping preferences, we believe that greater emphasis should be placed on understanding the social-psychological correlates play at the place-specific scale. Researchers have found that PPGIS can explore these correlates when used as part of a suite of methods (Elwood, 2010), including group valuation (Nijnik and Miller, 2017), visioning exercises (Ungar et al., 2020)d method (Author(s), XXXX). Additional social-psychological methods would be essential in uncovering the underlying discourses of concern (Buys et al., 2012; Etkin and Ho, 2007; Fløttum and Gjerstad, 2017). Similarly, exploring the various landscape value discourses would help support values-led management and planning (Albizua and Zografos, 2014; Rozema and Bond, 2015; Zografos and Martínez-Alier, 2009). Specifically, a multidisciplinary approach focused on social-psychological correlates can improve the framing of adaptation actions, thus bolstering a social license to act (Cradock-Henry et al., 2018).

Also beyond the scope of this paper was a comparative analysis between people’s spatial perception of risk and the projected climate risks (Brown et al., 2004). Since the local council has exclusively relied on projected risks for management and planning decisions, identifying areas where there is a high perception of risk and low projected risk, and vice versa, could assist local governments in reducing conflict between stakeholders and decision-makers (Carmen et al., 2022; Cunningham et al., 2023). Additionally, this paper discussed the role of the land management authority, which can limit the potential scope or ambition of local adaptation actions due to ambiguity or contestation over jurisdiction. Through a more comprehensive demonstration of the associations between management authority, landscape values, and risk perceptions, local councils could better communicate the challenges of adapting to climate change within existing policy frameworks (Hanssen et al., 2013; Measham et al., 2011).

This study identified the complex relationships between assigned landscape values, spatial perceptions of risk, and physical assets in the Huon Valley coastalscape. The measurement of the diverse community understandings of spatially-bounded values and risks can facilitate more effective climate change adaptation decision-making. We applied a similar methodological approach that Raymond and Brown (2011) established to participatory map community sentiments via the landscape value typology and select climate change risk variables. However, we expanded upon their methods by grounding our research in the psychological factors contributing to risk perceptions and the on-ground assets that anchor assigned values and risks. While the former demonstrated a need for future investigation, the latter provided place-specific evidence, leading to more accessible and actionable results for decision-makers who need to interpret where values are present and why they are valued in the first place. Furthermore, linking community knowledge of risks with the valued assets improves planners’ and decision-makers' ability to determine the appropriate adaptation and planning responses and operationalizes standard PPGIS methods. Equipped with this information, rural and regional councils can reframe their climate adaptation actions to get better stakeholder buy-in and, thus, better outcomes. When decision-making reflects the social and cultural understandings of values and risks and is grounded in the landscapes and physical assets in which climate change impacts manifest, climate change adaptation plans will ultimately be more collaborative, effective, and transformative.

Acknowledgements

We acknowledge the Melukerdee and Lyluequonny people who continue to be the traditional custodians of the land on which this research was undertaken. The authors are grateful for the financial support of the American Australia Association, the Centre for Marine Socioecology, and the University of Tasmania, which have made this research possible. The authors also thank the Huon Valley Council for their continued interest in the project and the community members who participated in the research. Lastly, we would like to acknowledge Rebecca M. B. Harris, who was integral in the conception and ideation of this project.

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

The authors have no relevant financial or non-financial interests to disclose.

All authors contributed to the study's conception and design. Material preparation, data collection and analysis were performed by Malcolm S. Johnson. The first draft of the manuscript was written by Malcolm S. Johnson, and all authors commented on and contributed to subsequent versions of the manuscript. All authors read and approved the final manuscript.

The datasets generated during and/or analysed during the current study are not publicly available due to ethical concerns surrounding anonymity but are available from the corresponding author upon reasonable request.

Abel, N., Gorddard, R., Harman, B., Leitch, A., Langridge, J., Ryan, A., Heyenga, S., 2011. Sea level rise, coastal development and planned retreat: analytical framework, governance principles and an Australian case study. Environmental science & policy 14, 279–288.
Adger, W.N., Barnett, J., Brown, K., Marshall, N., O’Brien, K., 2013. Cultural dimensions of climate change impacts and adaptation. Nature Clim Change 3, 112–117. https://doi.org/10.1038/nclimate1666
Albizua, A., Zografos, C., 2014. A Values-Based Approach to Vulnerability and Adaptation to Climate Change. Applying Q methodology in the Ebro Delta, Spain. Environmental Policy and Governance 24, 405–422. https://doi.org/10.1002/eet.1658
Artelle, K., Stephenson, J., Bragg, C., Housty, J., Housty, W., Kawharu, M., Turner, N., 2018. Values-led management: the guidance of place-based values in environmental relationships of the past, present, and future. Ecology and Society 23. https://doi.org/10.5751/ES-10357-230335
Australian Bureau of Statistics, 2021. 2021 Census data for Huon Valley.
Babelon, I., Ståhle, A., Balfors, B., 2017. Toward Cyborg PPGIS: exploring socio-technical requirements for the use of web-based PPGIS in two municipal planning cases, Stockholm region, Sweden. Journal of Environmental Planning and Management 60, 1366–1390. https://doi.org/10.1080/09640568.2016.1221798
Bernardo, F., 2013. Impact of place attachment on risk perception: Exploring the multidimensionality of risk and its magnitude. Studies in Psychology 34, 323–329. https://doi.org/10.1174/021093913808349253
Blennow, K., Persson, E., Persson, J., 2019. Are values related to culture, identity, community cohesion and sense of place the values most vulnerable to climate change? PLOS ONE 14, e0210426. https://doi.org/10.1371/journal.pone.0210426
Bonnes, M., Lee, T., 2017. Psychological theories for environmental issues. Routledge.
Booth, K., Harwood, A., 2016. Insurance as catastrophe: A geography of house and contents insurance in bushfire-prone places. Geoforum 69, 44–52. https://doi.org/10.1016/j.geoforum.2015.12.004
Boulton, E., 2016. Climate change as a ‘hyperobject’: a critical review of Timothy Morton’s reframing narrative. WIREs Climate Change 7, 772–785. https://doi.org/10.1002/wcc.410
Brandenburg, A.M., Carroll, M.S., 1995. Your place or mine?: The effect of place creation on environmental values and landscape meanings. Society & Natural Resources 8, 381–398.
Braunschweiger, D., Ingold, K., 2023. What drives local climate change adaptation? A qualitative comparative analysis. Environmental Science & Policy 145, 40–49. https://doi.org/10.1016/j.envsci.2023.03.013
Brechin, S.R., Wilshusen, P.R., Fortwangler, C.L., West, P.C., 2002. Beyond the Square Wheel: Toward a More Comprehensive Understanding of Biodiversity Conservation as Social and Political Process. Society & Natural Resources 15, 41–64. https://doi.org/10.1080/089419202317174011
Brown, G., 2017. A Review of Sampling Effects and Response Bias in Internet Participatory Mapping (PPGIS/PGIS/VGI). Transactions in GIS 21, 39–56. https://doi.org/10.1111/tgis.12207
Brown, G., Fagerholm, N., 2015. Empirical PPGIS/PGIS mapping of ecosystem services: A review and evaluation. Ecosystem Services, Best Practices for Mapping Ecosystem Services 13, 119–133. https://doi.org/10.1016/j.ecoser.2014.10.007
Brown, G., Kyttä, M., 2014. Key issues and research priorities for public participation GIS (PPGIS): A synthesis based on empirical research. Applied Geography 46, 122–136. https://doi.org/10.1016/j.apgeog.2013.11.004
Brown, G., Raymond, C., 2007. The relationship between place attachment and landscape values: Toward mapping place attachment. Applied Geography 27, 89–111. https://doi.org/10.1016/j.apgeog.2006.11.002
Brown, G., Raymond, C.M., Corcoran, J., 2015. Mapping and measuring place attachment. Applied Geography 57, 42–53. https://doi.org/10.1016/j.apgeog.2014.12.011
Brown, G., Smith, C., Alessa, L., Kliskey, A., 2004. A comparison of perceptions of biological value with scientific assessment of biological importance. Applied Geography 24, 161–180. https://doi.org/10.1016/j.apgeog.2004.03.006
Brown, G., Weber, D., 2012. Measuring change in place values using public participation GIS (PPGIS). Applied Geography 34, 316–324. https://doi.org/10.1016/j.apgeog.2011.12.007
Brown, T.C., 1984. The concept of value in resource allocation. Land economics 60, 231–246.
Buys, L., Miller, E., van Megen, K., 2012. Conceptualising climate change in rural Australia: community perceptions, attitudes and (in) actions. Regional Environmental Change 12, 237–248.
Carmen, E., Fazey, I., Ross, H., Bedinger, M., Smith, F.M., Prager, K., McClymont, K., Morrison, D., 2022. Building community resilience in a context of climate change: The role of social capital. Ambio 51, 1371–1387.
Cochrane, L., Corbett, J., 2020. Participatory mapping. Handbook of communication for development and social change 705–713.
Cradock-Henry, N.A., Frame, B., Preston, B.L., Reisinger, A., Rothman, D.S., 2018. Dynamic adaptive pathways in downscaled climate change scenarios. Climatic Change 150, 333–341. https://doi.org/10.1007/s10584-018-2270-7
Cunningham, C.A., Crick, H.Q.P., Morecroft, M.D., Thomas, C.D., Beale, C.M., 2023. Reconciling diverse viewpoints within systematic conservation planning. People and Nature 5, 621–632. https://doi.org/10.1002/pan3.10449
De Dominicis, S., Fornara, F., Ganucci Cancellieri, U., Twigger-Ross, C., Bonaiuto, M., 2015. We are at risk, and so what? Place attachment, environmental risk perceptions and preventive coping behaviours. Journal of Environmental Psychology 43, 66–78. https://doi.org/10.1016/j.jenvp.2015.05.010
Denny, L., Pisanu, N., 2019. Regional population trends in Tasmania: Issues and options.
Dillman, D.A., Smyth, J.D., Christian, L.M., 2014. Internet, phone, mail, and mixed-mode surveys: The tailored design method. John Wiley & Sons.
Dunlap, R.E., 2008. The New Environmental Paradigm Scale: From Marginality to Worldwide Use. The Journal of Environmental Education 40, 3–18. https://doi.org/10.3200/JOEE.40.1.3-18
Elwood, S., 2010. Mixed Methods: Thinking, Doing, and Asking in Multiple Ways, in: The SAGE Handbook of Qualitative Geography. SAGE Publications, Inc., 1 Oliver’s Yard, 55 City Road London EC1Y 1SP, pp. 94–114. https://doi.org/10.4135/9780857021090.n7
Etkin, D., Ho, E., 2007. Climate Change: Perceptions and Discourses of Risk. Journal of Risk Research 10, 623–641. https://doi.org/10.1080/13669870701281462
Fagerholm, N., Raymond, C.M., Olafsson, A.S., Brown, G., Rinne, T., Hasanzadeh, K., Broberg, A., Kyttä, M., 2021. A methodological framework for analysis of participatory mapping data in research, planning, and management. International Journal of Geographical Information Science 35, 1848–1875. https://doi.org/10.1080/13658816.2020.1869747
Fløttum, K., Gjerstad, Ø., 2017. Narratives in climate change discourse. WIREs Climate Change 8, e429. https://doi.org/10.1002/wcc.429
Frandsen, M., Paton, D., Sakariassen, K., 2011. Fostering community bushfire preparedness through engagement and empowerment. Australian Journal of Emergency Management, The 26, 23–30.
Grose, M., Barnes-Keoghan, I., Corney, S., White, C., Holz, G., Bennett, J., Gaynor, S., Bindoff, N., 2010. Climate Futures for Tasmania: General Climate Impacts Technical Report.
Hanssen, G.S., Mydske, P.K., Dahle, E., 2013. Multi-level coordination of climate change adaptation: by national hierarchical steering or by regional network governance? Local Environment 18, 869–887.
Hess, J.J., Malilay, J.N., Parkinson, A.J., 2008. Climate Change: The Importance of Place. American Journal of Preventive Medicine, Theme Issue: Climate Change and the Health of the Public 35, 468–478. https://doi.org/10.1016/j.amepre.2008.08.024
Huon Valley Council, 2023. Draft Huon Valley land use and development strategy. Huon Valley Council, Tasmania.
IPCC, 2022. Climate Change 2022: Impacts, Adaptation, and Vulnerability, Contribution of Working Group II to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge University Press, Cambridge, UK and New York, NY, USA.
King, N., Jones, A., 2021. An Analysis of the Potential for the Formation of ‘Nodes of Persisting Complexity.’ Sustainability 13, 8161. https://doi.org/10.3390/su13158161
Kruger, T.M., Beilin, R., 2014. A ‘responsibility for place’–firefighter deployment, local knowledge and risk. International journal of wildland fire 23, 577–584.
Lange, J., Gillespie, J., 2023. Bushfire, prescribed burning, and non-human protection. Geographical Research 61, 169–192. https://doi.org/10.1111/1745-5871.12558
Lyth, A., Harwood, A., Hobday, A.J., McDonald, J., 2016. Place influences in framing and understanding climate change adaptation challenges. Local Environment 21, 730–751.
McDonald, R.I., Chai, H.Y., Newell, B.R., 2015. Personal experience and the ‘psychological distance’ of climate change: An integrative review. Journal of Environmental Psychology 44, 109–118. https://doi.org/10.1016/j.jenvp.2015.10.003
Measham, T.G., Preston, B.L., Smith, T.F., Brooke, C., Gorddard, R., Withycombe, G., Morrison, C., 2011. Adapting to climate change through local municipal planning: barriers and challenges. Mitigation and adaptation strategies for global change 16, 889–909.
Mendel, L.C., Kirkpatrick, J.B., 2002. Historical Progress of Biodiversity Conservation in the Protected-Area System of Tasmania, Australia. Conservation Biology 16, 1520–1529. https://doi.org/10.1046/j.1523-1739.2002.01155.x
Morton, T., 2013. Hyperobjects: Philosophy and Ecology after the End of the World. U of Minnesota Press.
Moser, S.C., 2014. Communicating adaptation to climate change: the art and science of public engagement when climate change comes home. Wiley Interdisciplinary Reviews: Climate Change 5, 337–358.
Myers, N., Mittermeier, R.A., Mittermeier, C.G., Da Fonseca, G.A., Kent, J., 2000. Biodiversity hotspots for conservation priorities. Nature 403, 853–858.
Neale, T., 2018. ‘Are we wasting our time?’: bushfire practitioners and flammable futures in northern Australia. Social & Cultural Geography 19, 473–495.
Neale, T., Carter, R., Nelson, T., Bourke, M., 2019. Walking together: a decolonising experiment in bushfire management on Dja Dja Wurrung country. Cultural Geographies 26, 341–359.
Nijnik, M., Miller, D., 2017. Valuation of ecosystem services: paradox or Pandora’s box for decision-makers? One Ecosystem 2, e14808. https://doi.org/10.3897/oneeco.2.e14808
O’Neill, S., Nicholson-Cole, S., 2009. “Fear Won’t Do It”: Promoting Positive Engagement With Climate Change Through Visual and Iconic Representations. Science Communication 30, 355–379. https://doi.org/10.1177/1075547008329201
Osbaldiston, N., 2022. ‘The Summers Were Getting Hotter’: exploring motivations for migration to Tasmania away from mainland Australia. Australian Geographer 53, 461–476.
Persson, J., Sahlin, N.-E., Wallin, A., 2015. Climate change, values, and the cultural cognition thesis. Environmental Science & Policy 52, 1–5. https://doi.org/10.1016/j.envsci.2015.05.001
Peyton, K., Huber, G.A., Coppock, A., 2022. The generalizability of online experiments conducted during the COVID-19 pandemic. Journal of Experimental Political Science 9, 379–394.
Pidgeon, N., Hood, C., Jones, D., Turner, B., Gibson, R., 1992. Risk perception. Risk: Analysis, perception and management 89–134.
Praskievicz, S., 2022. Ground truth: Finding a “place” for climate change. Progress in Environmental Geography 1, 137–162. https://doi.org/10.1177/27539687221127035
Preston, B.L., Yuen, E.J., Westaway, R.M., 2011. Putting vulnerability to climate change on the map: a review of approaches, benefits, and risks. Sustainability science 6, 177–202.
Rands, M.R., Adams, W.M., Bennun, L., Butchart, S.H., Clements, A., Coomes, D., Entwistle, A., Hodge, I., Kapos, V., Scharlemann, J.P., 2010. Biodiversity conservation: challenges beyond 2010. science 329, 1298–1303.
Rawluk, A., Ford, R., Anderson, N., Williams, K., 2019. Exploring multiple dimensions of values and valuing: a conceptual framework for mapping and translating values for social-ecological research and practice. Sustain Sci 14, 1187–1200. https://doi.org/10.1007/s11625-018-0639-1
Raymond, C.M., Brown, G., 2011. Assessing spatial associations between perceptions of landscape value and climate change risk for use in climate change planning. Climatic Change 104, 653–678. https://doi.org/10.1007/s10584-010-9806-9
Raymond, C.M., Brown, G., Weber, D., 2010. The measurement of place attachment: Personal, community, and environmental connections. Journal of environmental psychology 30, 422–434.
Reid, K., Beilin, R., McLennan, J., 2020. Communities and responsibility: Narratives of place-identity in Australian bushfire landscapes. Geoforum 109, 35–43.
Remenyi, T.A., Earl, N., Love, P.T., Rollins, D.A., Harris, R.M.B., 2020. Climate Change Information for Decision Making. Huon Valley Council, Tasmania.
Rozema, J.G., Bond, A.J., 2015. Framing effectiveness in impact assessment: Discourse accommodation in controversial infrastructure development. Environmental Impact Assessment Review 50, 66–73.
Seymour, E., Curtis, A., Pannell, D., Allan, C., Roberts, A., 2010. Understanding the role of assigned values in natural resource management. Australasian Journal of Environmental Management 17, 142–153.
Singh, A.S., Zwickle, A., Bruskotter, J.T., Wilson, R., 2017. The perceived psychological distance of climate change impacts and its influence on support for adaptation policy. Environmental Science & Policy 73, 93–99.
Singh, S., Sagar, R., 2021. A critical look at online survey or questionnaire-based research studies during COVID-19. Asian Journal of Psychiatry 65, 102850. https://doi.org/10.1016/j.ajp.2021.102850
Ungar, M., McRuer, J., Liu, X., Theron, L., Blais, D., Schnurr, M., 2020. Social-ecological resilience through a biocultural lens: a participatory methodology to support global targets and local priorities. Ecology and Society 25. https://doi.org/10.5751/ES-11621-250308
van der Linden, S., 2015. The social-psychological determinants of climate change risk perceptions: Towards a comprehensive model. Journal of Environmental Psychology 41, 112–124. https://doi.org/10.1016/j.jenvp.2014.11.012
Wewerinke-Singh, M., Salili, D.H., 2020. Between negotiations and litigation: Vanuatu’s perspective on loss and damage from climate change. Climate Policy 20, 681–692.
Williams, D.R., Vaske, J.J., 2003. The measurement of place attachment: validity and generalizability of a psychometric approach. Forest Science. 49(6): 830–840.
Wilson, K.A., Underwood, E.C., Morrison, S.A., Klausmeyer, K.R., Murdoch, W.W., Reyers, B., Wardell-Johnson, G., Marquet, P.A., Rundel, P.W., McBride, M.F., 2007. Conserving biodiversity efficiently: what to do, where, and when. PLOS biology 5, e223.
Zografos, C., Martínez-Alier, J., 2009. The politics of landscape value: a case study of wind farm conflict in rural Catalonia. Environment and Planning A 41, 1726–1744.

Table 2 is not available with this version

IdentifyingvaluesandrisksSupMatv1.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Identifying relationships between landscape values, spatial perceptions of risk, and physical assets

Status:

Version 1

Abstract

Figures

Introduction

Background

Methods

3.1 Study location – Huon Valley coastalscape

3.2 Sampling technique – online mapping platform

3.3 Analysis techniques

3.3.1 Respondent profiles and connections to place

3.3.2 Attachment to place, risk perception, and scale interconnection

3.3.3 Statistical relationships between values and risks

3.3.4 Spatial distribution of landscape values and climate change risks

Results

4.1 Respondents

4.2 Social, experiential, and psychological factors

4.2.1 Scale reliability and correlations

4.2.2 Associations between scales, values, and risks

4.3 Correlations, associations, and relationships between values and risks

4.3.1 Relative importance of values and risks

4.3.2 Average nearest-neighbor

4.3.3 Statistical correlations

4.3.4 Spatial association between select values and risks

4.3.5 Spatial association of all mapped values and risks

Discussion

5.1 Influence of social-psychological correlates

5.2 Correlation between values and risk

5.3 Spatial associations of select risks and values

5.4 Implications for adaptation planning

5.5 Future research opportunities

Conclusion

Declarations

References

Table 2

Supplementary Files

Status:

Version 1