Edited by: Maria Grazia Pennino, Oceanographic Center of Vigo, Spanish Institute of Oceanography, Spain
Reviewed by: Blaine D. Griffen, Brigham Young University, United States; Kevin Alexander Hovel, San Diego State University, United States
This article was submitted to Marine Conservation and Sustainability, a section of the journal Frontiers in Marine Science
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Dungeness crab (
Fisheries of the United States (U.S.) west coast are coupled natural-human systems that cross state borders, and integrate private and public sectors. They involve complex relationships between coastal socio-economics, resource management, and environmental factors. The integral role of humans in fishery systems has been formally recognized with a shift toward ecosystem-based management in natural resource programs (
The consequences of changing ocean conditions due to climate change are already being experienced across fishery systems (
Therefore, we aimed to elucidate the impact of changing ocean conditions on coastal communities through the lens of a socio-economically important fishery in the CCS. In this study, we focused on commercial catch of Dungeness crab (
Changes in ocean temperatures and currents may impact survival and dispersal of Dungeness crab larvae in the CCS, and subsequently the magnitude and distribution of commercial catch along the U.S. west coast. The 3- to 4-month pelagic larval stage of Dungeness crab is the most vulnerable life stage, and the crabs become increasingly resilient to environmental factors, like temperature, as they develop into adults (
Management of Dungeness crab is atypical compared to other U.S. west coast fisheries because total allowable catch determinations are not based on calculations for sustainable yield and do not incorporate a formal stock assessment (
Using an interdisciplinary framework, we investigated how changing ocean conditions affect commercial catch of Dungeness crab and associated fishery communities of the U.S. west coast. Through the process of four steps, our analysis utilized both existing and novel metrics for exposure and susceptibility in the Dungeness crab coupled natural-human system. We defined the potential “exposure” of fishing communities to altered Dungeness crab catch due to changing ocean conditions via the first two steps. First, (1) we determined the spatial and temporal relationship between various ocean conditions and Dungeness crab commercial catch along the CCS from 1981 to 2017. Then, (2) this relationship was used to examine the magnitude of change in Dungeness crab catch for U.S. west coast port communities resulting from two ocean condition scenarios. Next, (3) we assessed the degree to which changes in Dungeness crab commercial catch may impact 18 fishery-reliant CCS communities (
Map of the 18 major U.S. west coast port communities considered in this study within the Washington, Oregon, Northern California, and Central California management units. See
Data were obtained from multiple state, federal, and academic sources (
We gathered historical Dungeness crab commercial catch data (1981–2017) within the CCS for major U.S. west coast fishing ports (two Washington regions, seven Oregon ports, and seven California ports) (
A time series of the ocean conditions occurring within or affecting the CCS was compiled for 1977–2017 based on our hypotheses about the impact of ocean conditions on Dungeness crab survival and dispersal (
List of ocean conditions analyzed for their relationship to Dungeness crab commercial catch per unit effort (CPUE) across the U.S. west coast, description of how the ocean condition value was summarized for analysis, and its hypothesized effect on CPUE.
Ocean condition | Summary value | Hypothesized effect: |
Pacific Decadal Oscillation (PDO) | Sum of January–June monthly averages | Stronger negative PDO ( |
North Pacific Gyre Oscillation (NPGO) | Sum of January–June monthly averages | |
Multivariate El Niño Southern Oscillation Index (MEI) | Sum of January–June monthly averages | |
Southern Oscillation Index (SOI) | Sum of January-June monthly averages | |
Sea Surface Temperature (SST) | Average January–June (2° latitude increments from 36°N to 48°N) | Lower SST ( |
Upwelling (Upw) | Daily sum from April-August (3° latitude increments from 36°N to 48°N) | Stronger Upw ( |
Spring Transition Index (STI) | Day of year (3° latitude increments from 36°N to 48°N) | Earlier STI ( |
Northward component of Ekman Transport (EkTrans) | Sum of January-June monthly averages (3° latitude increments from 36°N to 48°N) | More negative northward component of EkTrans ( |
Ocean conditions measurements from 1977 to 2017. Regional factors (PDO, ENSO SOI, ENSO MEI, and NPGO) are monthly index values. Local factors (EkTrans, SST, Upw, and STI) are shown, for example, at 45°N. Northward EkTrans (value × 10 = metric tons second–1 kilometer–1) are shown as monthly values. SST (°C), Upw (meters3 second–1 per 100 m of coastline), and STI (day of year) are annual values.
Generalized additive mixed models (GAMM) were used to assess the spatial and temporal relationship between ocean conditions and commercial Dungeness crab CPUE along the U.S. west coast. To allow for the spatially variable effects of ocean conditions on Dungeness crab CPUE and to account for simultaneously testing the effects of ocean conditions at multiple locations, we applied a variable coefficient GAMM (VC_GAMM) framework. The VC_GAMM allows for variable model intercepts and variable linear relationships between the response and the design variable (
The GAMM and VC_GAMM were built using the R packages ‘mgcv’ (v.1.8-28;
Where
The best model for each ocean condition was chosen based on LR tests and AIC. The model with the lowest AIC value was chosen as the best model if it was the reduced (simpler) model. However, if the model with the lowest AIC value was the more complex model, it had to be significantly different (LR test,
The CMIP5 multi-model ensemble of the International Panel on Climate Change (IPCC) Fifth Assessment Report (AR5) predicts that SST will increase by 0.3°C decade–1 through the end of the century (
Possible future (2080) sea surface temperature (SST) scenarios for the California Current System (CCS) based on the IPCC AR5.
Moderate | RCP 4.5 W/m2 | +1.7°C |
High | RCP 8.5 W/m2 | + 2.8°C |
To determine the impact of a change in Dungeness crab CPUE (exposure) on fishery dependent coastal communities along the CCS, we first assessed the current importance of Dungeness crab to these communities by utilizing existing indices for reliance and vulnerability. The National Oceanic and Atmospheric Administration designed frameworks to evaluate fishery reliance and engagement of coastal communities (
A multi-point social vulnerability analysis in 2014 provided information on the current overall socio-economic stability of each coastal community (
To determine the relative degree to which a change in Dungeness crab CPUE could impact coastal communities, we combined Dungeness crab-specific fishery reliance indices with CSVI into a novel “susceptibility” score. Our susceptibility score for each community was calculated as the product of CSVI and reliance. CSVI encapsulates the relative ability of communities to weather hardships, where lower indices indicate a community that is less vulnerable in the face of an unwanted event due to its social well-being. The reliance index denotes the per capita presence of the Dungeness crab fishery in the community relative to other communities, which expresses how important the fishery is to the economic and social well-being of the community. The value of the product, susceptibility, was used to represent the vulnerability of a community, scaled by reliance on Dungeness crab and is a quantitative representation of the degree to which each community would be influenced by a change in Dungeness crab CPUE. Communities that are both heavily reliant on Dungeness crab and socially vulnerable would have high susceptibility scores, highlighting the scaling effect of these two properties. In contrast, communities with either high vulnerability and low reliance on Dungeness crab, or low vulnerability and high reliance on crab would have similar susceptibility scores. Finally, a community with both low reliance and vulnerability, would be the least susceptible because they are generally resilient to change, and a change in the Dungeness crab fishery would not have a large impact on the community. The compounded risk for each community was examined by comparing the exposure to a change in Dungeness crab CPUE under increased SST scenarios with a region-specific susceptibility score. Although exact terminology differs, similar approaches have been adopted in other recent studies examining the exposure-vulnerability of coastal communities to changing ocean conditions (e.g.,
From 1981 to 2017, Dungeness crab commercial catch within 16 major ports/regions of the CCS varied annually (total U.S. west coast mean catch = 985,183 kg;
Total commercial catch of Dungeness crab in kilograms (kg) per year (solid line) and catch per unit effort (dashed line) from 1981 to 2017 for the U.S. west coast, including 16 ports/regions in Washington, Oregon, and California. Catch per unit effort (CPUE) was calculated as kg caught per number of fish tickets per year. Data were provided by WDFW 2018, ODFW 2018, and CDFW 2018.
Percent of total Dungeness crab commercial catch per unit effort (CPUE) landed annually in 16 U.S. west coast ports from 1981 to 2017. Warm colors indicate a high percentage of the total annual CPUE and cool colors indicate a low percentage of the total annual CPUE. Total annual CPUE was calculated as kilograms caught per number of fish tickets per year. Ports correspond to those shown in
The results of the stationary GAMM and non-stationary VC_GAMM and selected models are listed in
Selected models of individual ocean conditions lagged 3, 4, or 5 years and Dungeness crab commercial catch per unit effort ln(CPUE) along the U.S. west coast, including the adjusted
Pacific Decadal Oscillation (PDO) | 0.54 | 0 | |
North Pacific Gyre Oscillation (NPGO) | 0.50 | 19 | |
Multivariate El Niño Southern Oscillation Index (MEI) | 0.46 | 39 | |
Southern Oscillation Index (SOI) | 0.47 | 39 | |
Sea Surface Temperature (SST) | 0.51 | 49 | |
Upwelling (Upw) | 0.43 | 43 | |
Spring Transition Index (STI) | 0.42 | 48 | |
Northward component of Ekman Transport (EkTrans) | 0.45 | 31 |
There was support (
Consistent with other temperature-related oceanic indices (e.g., PDO), there was support (
Coefficients of SST lagged by
There was support (
Predictions for a +1.7 or +2.8°C increase in CCS SST indicated that Dungeness crab ln(CPUE) along the U.S. west coast would decrease under these two scenarios (
Because paired CPUE data and socio-economic indices (reliance and CSVI) were not available for all 18 selected coastal communities, we divided them into four regions that reflect the U.S. west coast fishery management divisions: Washington, Oregon, Northern California, and Central California. Compared to all communities across the CCS, Westport, Washington was an outlier with a very high Dungeness crab reliance index (
Reliance index, community social vulnerability index (CSVI), susceptibility score, and predicted percent decrease in ln(CPUE) by port and by region (
Dungeness crab reliance index for 2015 versus social vulnerability (CSVI) for 2014. Reliance scores consider per capita engagement in the Dungeness crab fishing industry, while CSVI indicates the overall vulnerability of a community. This figure excludes Westport, WA, which had anomalously high reliance and moderate vulnerability (
The calculated susceptibility score (product of reliance and CSVI) and the percent decrease in Dungeness crab CPUE under future SST scenarios shows that the overall risk of a shift in Dungeness crab CPUE is moderate to high across the CCS, depending on the level of SST warming (
Predictions from the SST VC_GAMM indicated that the northern portion of the CCS Dungeness crab fishery is the least exposed to a change in CPUE due to changes in SST (32.31–50.08% decrease in CPUE, depending on latitude and SST scenario;
The SST VC_GAMM predicted higher exposure of Oregon Dungeness crab CPUE to SST changes than for Washington (32.05–69.07% decrease in CPUE, depending on latitude and SST scenario;
Northern California port communities showed higher exposure to changes in CPUE predicted by the SST VC_GAMM than Washington or Oregon (51.64–79.62% decrease in CPUE, depending on latitude and SST scenario;
The southernmost Dungeness crab ports had the greatest exposure to CPUE changes due to SST increases, according to the VC_GAMM (62.06–100.00% decrease in CPUE, depending on latitude and future SST scenario;
The impacts of climate change on marine systems will alter ecosystem-scale functions including productivity, food web and disease dynamics, and species distributions (
Statistical analysis of Dungeness crab CPUE and ocean conditions identified a mix of relationships that were lagged by 3–5 years. This finding upholds previous work and confirms the importance of favorable ocean conditions, such as negative PDO and positive SOI indices, for promoting larval crab survival and transport in the CCS (
Overall, we found that the models explaining the largest amount of variance were those with PDO, SST, and NPGO, individually. Consistent with our findings, negative PDO conditions during the vulnerable early life history stages of Dungeness crab, and other exploited marine fish, have been shown to increase survival to harvestable size (
While we focus on the SST model for the purpose of determining exposure of CPUE to changing ocean conditions, it is important to recognize that SST is closely correlated with other climate-related phenomena. For example, increasing SST is likely to increase occurrences of hypoxic events (
Our analysis showed that increasing SST in the CCS would decrease U.S. west coast CPUE of Dungeness crab with larger decreases in CPUE projected for the RCP 8.5 W/m2 radiative forcing scenario (+2.8°C) compared to RCP 4.5 W/m2 (+1.7°C). Decreases in CPUE would likely be caused by a shift in offshore distribution and accessibility of Dungeness crab to the fishing industry. However, the decrease in CPUE may not be monotonic along the coast, as the best model included a non-stationary effect of SST at a 5-year lag. Under both warming scenarios, our model found greater exposure of Dungeness crab CPUE to SST for southern latitude ports than northern latitude ports and indicates the possibility for a northward distribution shift of U.S. west coast Dungeness crab. Furthermore, there may be regional differences in the rate of SST warming, which could accentuate a northward distribution shift.
There is a generally good consensus that SST will increase in the CCS (
Economically valuable fisheries provide a direct link between changing ocean conditions and communities. U.S. west coast communities showed varying levels of social vulnerability and reliance on the Dungeness crab fishery that did not necessarily align with regional or management boundaries. Combining these metrics into a susceptibility score allowed us to determine that the risk of a change in Dungeness crab distribution or availability is unequal, with generally greater consequences for northern communities.
Given the projected decrease in Dungeness crab catch with increased SST, communities with the greatest potential for exposure were those in the central California management region. Although the likelihood of communities such as Morro Bay and Monterey experiencing a reduction in catch of Dungeness crab was relatively high, their overall susceptibility was low. On the other hand, Bodega Bay had higher reliance on the Dungeness crab fishery and was the most susceptible of the central California communities.
Oregon and Washington were less exposed to changes in Dungeness crab catch due to increased SST compared to California. However, many of these communities showed high social vulnerability and, ultimately, higher susceptibility. Therefore, the risk of reduced Dungeness crab catch would likely be more detrimental to communities in Oregon and Washington. As noted above, Bodega Bay, California represents a potential exception to this pattern. In addition, communities across the CCS have already been impacted by other climate-related phenomena and the subsequent management responses, including closures for the presence of domoic acid in crabs and hypoxia-related mass mortality of adult crab. Expected increases in the potency, duration, and occurrence rate of harmful algal blooms in marine systems (
One caveat of this discussion is the fact that CSVI captures overall vulnerability of the “community of place” and is not specific to participants in the commercial fishing industry, which represents the “community of interest” (
In general, it is analytically challenging to capture the effect of individual ocean conditions on marine species due to the myriad of interacting physical and biological factors, variability in ocean conditions at multiple spatial and temporal scales, and often limited data on the distribution and abundance. Given these complexities and limited data on certain important ocean conditions, our modeling approach allowed us to describe historical trends and project CPUE under increased SST scenarios, but lacks mechanistic explanation for these changes. In the absence of mechanistic understanding, our projected future changes in CPUE with increased SST are based on the assumption that underlying mechanisms are linear and will continue unchanged into the future. However, synergies between temperature, ocean acidification, and hypoxia have been shown to alter the thermal tolerance of marine species (
For this analysis, we had to limit our inference to that of Dungeness crab CPUE due to the lack of available information about the actual offshore abundance and distribution of crab. Management of Dungeness crab does not include a formal stock assessment and adequate spatial information captured in fishing vessel logbooks is not available for all regions along the U.S. west coast. Without consistent logbook information, it is difficult to estimate fishing effort across all three states. Therefore, we standardized catch across ports and years using fish tickets following
Although our data covered 36 years, some ocean condition factors operate on longer time scales. For instance, the data only spanned two major oscillations in PDO, which may have limited our ability to effectively describe the relationship between PDO and CPUE. Other ocean conditions not included in this analysis, due to insufficient data, may also contribute to Dungeness crab survival and distribution along the U.S. west coast, including hypoxia and ocean acidification. Such data could improve our mechanistic understanding of changes in Dungeness crab CPUE. These limitations reiterate the value of collecting long-term data series related to the success of socio-economically important fisheries under climate change.
Socio-economic data in our analysis were temporally limited and represent only a snapshot of each community. Indices obtained were based on 2014 (CSVI) and 2015 (Dungeness crab reliance) and may not be a comprehensive representation of the community condition from 1981 to 2017; however, these data represent warm ocean years in the CCS that may be common in the future (
Climate change is creating scenarios that are environmentally, ecologically, and socially novel. We found that decreased Dungeness crab catch from increasing ocean temperature has the potential to greatly impact fishery participants and U.S. west coast communities. In addition, substantial risks to the Dungeness crab fishery exist from other climate-related phenomena (e.g., domoic acid, hypoxia, and ocean acidification) that often co-occur with temperature. This study provides a valuable framework for understanding the social and ecological impacts of one important fishery in a single analysis, however, we recognize that this examination is insufficient to capture the full risk of climate change on U.S. west coast communities. For instance, many exposure-vulnerability analyses suffer from limited understanding of social and ecological adaptive capacity, particularly the timescales over which they operate. The social-ecological context of marine fisheries provides a particularly promising avenue for exploring adaptive capacity and untangling the complexity of coupled natural-human systems. There is great potential for future work to understand the possibility for resilience and adaptation in both coastal communities and target species. Among other social factors that contribute to adaptive capacity, flexibility to change strategies (
The datasets generated for this study are available on request to the corresponding author.
All authors contributed to the conceptualization and design of the study. CM and EL obtained the fishery and environmental data and performed the analysis. AS and AJ obtained the social data and performed the analysis. CM, EL, AS, KS, and AJ wrote and revised the manuscript.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Engagement with numerous stakeholders and scientists was necessary for the success of this project. Data and guidance were provided by Kelly Corbett (ODFW), Troy Buell (ODFW), Mitch Vance (ODFW), Matt Hunter (ODFW), Thomas Swearingen (ODFW), Christy Jahasz (CDFW), Daniel Ayers (WDFW), Karma Norman (NOAA), Anna Varney (PSMFC), Jack Barth (OSU), Jennifer Fisher (NOAA), Gil Sylvia (OSU), Chris Cusack (OSU), and Malin Pinsky (Rutgers). Special thanks to the Oregon State University National Science Foundation Research Traineeship staff and faculty for supporting this project. Feedback from Francis Chan, Flaxen Conway, Sally Hacker, David Koslicki, Kathleen O’Malley, Su Sponaugle, and Lorenzo Ciannelli helped guide this research toward a productive and meaningful outcome.
The Supplementary Material for this article can be found online at: