Fig 1.
Perceptions of vulnerability framework.
Perceived vulnerability of fishers was assessed by combining perceived exposure, sensitivity, and adaptive capacity. Design by SJ Bowden.
Table 1.
Concern themes and the questions that contributed towards them for the cluster analysis.
Fig 2.
a) Number of responses received by fishery and b) industry role. c) Response numbers by county of homeport and regions regularly fished. Individuals could respond multiple times for each question except for homeport. Fishing regions were defined as: Puget Sound and the Strait of Juan de Fuca; WA Coast–Cape Flattery, WA to Cape Disappointment, WA; the Columbia River; OR Coast–South of the Columbia River to Mendocino, CA; Northern CA–Mendocino, CA to Point Reyes, CA; Central CA–Point Reyes, CA to Point Conception, CA; Southern CA–south of Point Conception, CA. The map was made using the tigris R package which uses TIGER/Line shapefiles from the US Census Bureau. https://www.census.gov/geographies/mapping-files/time-series/geo/tiger-line-file.html.
Table 2.
Summary statistics of survey respondents.
One individual that lives in Alaska responded, they fish in Washington and were included in the Washington group during analysis to preserve anonymity.
Fig 3.
Future of fishing and climate change.
The percentage of respondents that agreed or disagreed with the statements on the left of the figure. Numbers on the far right and far left are the sum of the strongly and somewhat agree or disagree responses and those in the middle are the percentage of people who answered neutral for that statement. n = 162.
Table 3.
Common themes and observations from open-ended questions about range and timing shifts, and effects of climate change on fishing.
Fig 4.
Heatmap of dimensions of vulnerability by fishery.
Relative average levels of exposure, sensitivity, adaptive capacity, and vulnerability for select fisheries where darker red indicates a higher contribution to vulnerability. Each column is scaled independent of the others, therefore the score associated with the darkest red in exposure is not equivalent to that same color in the vulnerability column.
Fig 5.
Density curves of components of vulnerability.
The distribution of scores within each component of vulnerability is shown with respondents grouped by their responses to the prompt “I believe climate change is occurring.” Letters indicate which groups are statistically different (p < 0.05) according to the results of pairwise Wilcoxon rank sum tests with Bonferroni correction.
Table 4.
Mean scores of components of vulnerability when people are clustered by their belief that climate change is occurring, and results of the Kruskal-Wallis statistical test for each component.
Fig 6.
Boxplot of average scores for concern clusters 1, 2, and 3 across the six themes.
The line in the middle of the box is the median score, the upper and lower end of the box indicate the 75% and 25% percentiles respectively, and dots represent potential outliers. Because of the way the outlook statements were phrased, a higher score indicates a more negative outlook for future fishing prospects. Letters indicate significantly different groups p < 0.05 according to the results of pairwise t-tests. There are 65 people in cluster 1, 60 in cluster 2, and 37 in cluster 3.
Fig 7.
Perceived vulnerability of individuals.
Colors correspond to the clustered concern groups and black lines correspond to median risk and adaptive capacity. Quadrant vulnerability profiles are 1. Potential adapters, 2. Higher concern, 3. High latent risk, and 4. Lower concern [77]. X axis is oriented from 1 to 0 because higher adaptive capacity reduces vulnerability.