1 Introduction: Cape Falcon Hook and Line Biology & Oceanography Report

Data from hook and line (HnL) surveys can be used to explore questions about fish abundance and size from a fishery-independent monitoring method. Oceanographic data that are collected near HnL sampling sites provide the opportunity to understand local oceanographic processes that may influence the performance of hook and line surveys to represent the fish community. We analyzed HnL data together with co-located and nearby oceanographic data to assess the influence of time-varying environmental conditions on the responses of fish communities as represented by hook and line sampling. Specifically, we assess the effect of bottom dissolved oxygen, wind stress (as a proxy for upwelling), and bottom temperature on the aggregate catch and biomass rates of all species, as well as the catch and biomass rates and size for select focal species (for more information about focal species, see Focal Species Documentation).

1.1 Research Questions

Do oxygen levels influence fish responses as gathered with hook and line data ?
Does wind stress (as a proxy for upwelling) influence fish responses as gathered with hook and line data?
Does mean bottom temperature influence fish responses as gathered with hook and line data?

For each of the above research questions, we assessed fish responses to oceanographic conditions in aggregate and individually for each focal species. Aggregate fish responses with hook and line data include catch and biomass rates (CPUE and BPUE). Focal species responses with hook and line data include catch and biomass rates (CPUE and BPUE), and mean size.

We have the following hypotheses:

Low oxygen (hypoxic) conditions will lead to lower catch and biomass rates (CPUE and BPUE) with hook and line gear.
Stronger upwelling conditions (negative wind stress) will lead to higher catch and biomass rates (CPUE and BPUE) with hook and line gear.
There are two possible hypotheses related to changes in bottom temperatures. The first is that warmer bottom temperature will lead to higher catch and biomass rates, or larger sizes, with hook and line gear, as documented for when warmer bottom temperatures influence metabolic rates. Alternatively, colder bottom temperatures can be tightly coupled with coastal upwelling, and therefore colder temperature could lead to higher catch and biomass rates, with warmer temperatures associated with lower catch and biomass rates (Bakun et al 2015).

2 Takeaways

We now review the results of our analyses in light of our research questions and summarize what they tell us in a broader context about the Cape Falcon Marine Reserve.

We did not find any evidence of fish response to dissolved oxygen levels.

We hypothesized that at low dissolved oxygen levels, we would see a response in CPUE, BPUE, and fish size either in aggregate or at a species-specific level. We observed a limited range of dissolved oxygen values at Cape Falcon Marine Reserve and at the associated comparison areas; however, we did not observe hypoxic waters. This may explain the lack of rish response to this oceanographic parameter.

Aggregate BPUE, and Lingcod CPUE and BPUE, were elevated when upwelling occurred the day prior to HnL sampling.

This finding was in alignment with our hypothesis: that stronger upwelling conditions, either on the day of or day prior to sampling, could lead to higher CPUE or BPUE. However, Black Rockfish did not show any response to upwelling conditions. No relationships were found with the proxy for upwelling (wind-stress) on the day of HnL sampling for any fish responses.

We did not find any evidence of a fish response to bottom temperature.

We observed a wide range of bottom temperatures, with anomalously high temperatures in the fall of 2019 (max. temp: 15.1C) indicating that we likely captured the effect of a marine heatwave (Amaya t al., 2020). It is possible that focal species were affected by cumulative heat stress that is not well captured by the day of and day prior window for temperature records; however, our analysis did not account for potentially higher order dynamics.

Conducting regular, co-located oceanographic and biological monitoring allowed us to identify biological responses to ocean changes over several years.

Co-located oceanographic and biological data time series increase our understanding of biological responses to changes in ocean conditions over various time scales. In addition, capturing inter-seasonal and inter-annual variability is critical to better understand biological thresholds and the effect of multiple stressors in a changing ocean.

3 Methods

3.1 Hook and Line Surveys

Hook and Line (HnL) surveys were conducted in the Cape Falcon Marine Reserve, and its three comparison areas representing different fishing pressures - low (Cape Meares), moderate (Manzanita and Nehalem Reefs, Dinner Plate Reef, Castle Rock Reef), and high (Three Arch Rocks). Surveys began in 2014 with unequal survey effort. In the initial years there was a strong focus to place more survey effort in the reserve to ensure adequate characterization of baseline conditions prior to closure. Survey effort targeted 2 days in the marine reserve and Moderate Fishing Pressure Comparison Area, and 2 days at the Low and High Fishing Pressure Comparison Areas for both spring and fall surveys. HnL data were collected for four years: 2014, 2015, 2017 and 2019.

Each day between five to six cells (500m x 500 m grids) are targeted, with 3 fifteen-minute fishing drifts occurring in each cell. All data collected from drifts are combined for a single cell; the unit of replication for hook and line surveys is at the cell-day level (except for size, which is at the individual fish level). All fish caught during each drift are identified to species, measured and released. During each drift we also determined the amount of time not spent fishing, such as when anglers hang up on the bottom, stop to rest, or take a photo with their fish caught. This time is factored into the total time spent fishing, to better represent fishing effort during these surveys. We then calculated both a catch and biomass per unit effort (CPUE and BPUE) and mean size (length, cm) for each given cell-day and species. For additional details on data collection, please review documentation in the Methods Appendix.

Focal Species

There are six focal fish species for the OR Marine Reserves Ecological Monitoring Program:

Black Rockfish; Sebastes melanops
Blue/Deacon Rockfish; Sebastes mystinus / S. diaconus
China Rockfish; Sebastes nebulosus
Yelloweye Rockfish; Sebastes ruberrimus
Cabezon; Scorpaenichthys marmoratus
Lingcod; Ophiodon elongatus

These species were chosen based on their ecological, economic or management importance. For more information please refer to the Methods Appendix detailing focal species selection.

3.2 Oceanographic data collection

Monitoring oceanographic conditions at the Cape Falcon Marine Reserve began in 2019 – with moorings placed in the marine reserve and Low Fishing Pressure Comparison Area. These moorings collected data on temperature, salinity and oxygen in shallow (15 meters) nearshore waters. Additionally in 2019, crab pots with dissolved oxygen and temperature data were also deployed on hook and line sampling days (n=7). To fill in oceanographic variables during sampling years prior to oceanographic monitoring, we used available data from oceanographic sensors in Tillamook Bay (NOAA’s National Data Buoy Center Station TLB03, https://www.ndbc.noaa.gov/station_page.php?station=tlbo3), the closest geographical data available.

North-South wind stress is a well-known proxy for upwelling (e.g., Allen, 1980, Samelson et al., 2002), where more negative values indicate more upwelling. Wind stress was estimated using data for each HnL sampling date for four years (2014, 2015, 2017, 2019) from Newport (NOAA’s National Data Buoy Center Station NWP03, https://www.ndbc.noaa.gov/station_page.php?station=nwpo3). Wind stress data were also estimated from one day prior to each HnL sampling date to represent conditions on the day prior to sampling.

Bottom temperature data were collated for sampling days in 2014, 2015, 2017 and 2019. For 2014-2017, in the absence of in situ temperature records, temperature was estimated from the NDBC sensor in Tillamook Bay and regressed to calculate temperature values near HnL sampling areas at 15m depth. Bottom temperature data were also collected during 2019 using sensors deployed inside crab pots or mooring data.

3.3 Analysis methods

Levels of data aggregation

We aggregated oceanographic variables by day, such that we have only one value per variable per day. We aggregated biological data to two different levels. Data were aggregated by day (one value per day across all survey sites) and by site-day (one value per site per day). Oceanographic and biological response data were joined based on the common value of date. if there was more than one area sampled in a given day (e.g., Cape Falcon and Moderate Fishing Pressure Comparison Area both sampled on 5/15/2014), we repeated the value of the oceanographic variable for both areas because we only have one value per oceanographic variable per day.

Statistical analyses

We used correlation analyses to assess the relationship between oceanographic variables and biological response variables. We used the Spearman’s rho correlation coefficient rather than the Pearson correlation coefficient because our sample size was limited in some cases. Spearman’s rho is a nonparametric correlation coefficient based on ranked variables. Spearman’s rho is robust to small sample sizes, outliers, and violations of the assumption of normally distributed variables. When assessing single-species biological responses, due to sample size concerns, we conducted correlation analyses for only the two most abundant species, Black Rockfish and Lingcod. This decision was in alignment with the decision to restrict modeling efforts with the HnL data set to the same two species.

4 Cape Falcon Marine Reserve Results

4.1 Dissolved Oxygen

Figure 1. Dissolved oxygen values for each of the seven HnL sampling days in 2019. The hypoxia threshold (1.4ml/L; Rabalais et al., 2010) is indicated by the dashed line.

4.1.1 Aggregate Data

Data are shown aggregated at different levels, including:

By day (n=7)
By site-day (one value per site per day, n=15)

4.1.1.1 By day

Figure 2. Mean aggregate CPUE and BPUE by dissolved oxygen value. Each data point represents one day. Spearman’s rank correlation coefficient is reported.

4.1.1.2 By site-day

Figure 3. Mean aggregate CPUE and BPUE by dissolved oxygen value. Each data point represents one site-day. Spearman’s rank correlation coefficient is reported.

There was no apparent trend in aggregate CPUE or BPUE with dissolved oxygen values observed at the Cape Falcon Marine Reserve or associated comparison areas.

There were no statistically significant correlations between mean CPUE or BPUE, aggregated to the day or site-day level, and bottom temperature on the day of (or day prior to) HnL sampling.

4.1.2 Focal Species

Mean CPUE, BPUE and size are plotted below as a function of dissolved oxygen for both Black Rockfish and Lingcod.

Figure 4. Mean CPUE, BPUE, and length by dissolved oxygen value for two focal species, Black Rockfish and Lingcod. Each data point represents one site-day. Spearman’s rank correlation coefficient is reported.

We did not find any statistically significant correlations between the CPUE, BPUE, or mean size and dissolved oxygen in either Black Rockfish or Lingcod.

We did not find any statistically significant correlations between the CPUE, BPUE, or average size and dissolved oxygen in either Black Rockfish or Lingcod.
The four other focal species – Blue/Deacon Rockfish, China Rockfish, Yelloweye Rockfish and Cabezon – had low catch rates across all sites – and therefore had sample sizes too low to statistically analyze and no apparent trends were seen with the data available.

Figure 5. Mean CPUE, BPUE, and length by dissolved oxygen value for all focal species. Each data point represents one site-day.

4.2 North-South Wind Stress (Upwelling)

Figure 6. N-S wind stress (black) and lagged N-S wind stress (blue) for each of the HnL sampling days for 2014, 2015, 2017 & 2019.

4.2.1 Aggregate Data

Data are shown aggregated at different levels, including:

By day (one value per day, n=26)
By site-day (one value per site per day, n=51)

4.2.1.1 By day

Figure 7. Mean aggregate CPUE and BPUE by N-S wind stress and lagged N-S wind stress. Negative N-S wind stress values (to the left of the vertical line) indicate upwelling. Each data point represents one day. Spearman’s rank correlation coefficient is reported.

4.2.1.2 By site-day

Figure 8. Mean aggregate CPUE and BPUE by N-S wind stress and lagged N-S wind stress. Negative N-S wind stress values (to the left of the vertical line) indicate upwelling. Each data point represents one area-day. Spearman’s rank correlation coefficient is reported.

Looking across sampling days, upwelling on the day prior to HnL sampling was significantly related to increased BPUE. There were no apparent trends between upwelling and aggregate CPUE.

There was significant negative correlation (p < 0.05) between mean aggregate BPUE and wind stress on the day prior to HnL sampling. There were no significant correlations between CPUE and wind stress on the day of (or day prior to) HnL sampling. Additionally, there were no significant correlations between CPUE or BPUE and wind stress on the day of (or day prior to) HnL sampling when data were aggregated by site-day.

4.2.2 Focal Species

Mean CPUE, BPUE and size are plotted below as a function of wind stress on the day of (and day prior) to HnL sampling for both Black Rockfish and Lingcod.

4.2.2.1 By day

Figure 9. Mean CPUE, BPUE, and length by N-S wind stress (A) and lagged N-S wind stress (B) for two focal species, Black Rockfish and Lingcod. Negative N-S wind stress values (to the left of the vertical line) indicate upwelling. Each data point represents one day. Spearman’s rank correlation coefficient is reported.

4.2.2.2 By site-day

Figure 10. Mean CPUE, BPUE, and length by N-S wind stress (A) and lagged N-S wind stress (B) for two focal species, Black Rockfish and Lingcod. Negative N-S wind stress values (to the left of the vertical line) indicate upwelling. Each data point represents one area-day. Spearman’s rank correlation coefficient is reported.

Lingcod CPUE and BPUE were increased when upwelling occurred during (and the day prior to) HnL sampling.

When looking across sampling days, there was evidence (p < 0.05) of a negative correlation between Lingcod CPUE and N-S wind stress, where more negative values of wind stress (indicating upwelling) on the day of HnL sampling were associated with increased Lingcod CPUE (Figure 9A, column 1). However, the removal of an apparent outlier (date: 2017-05-22, site: moderate fishing pressure) rendered this pattern not statistically significant (p>0.05). There was evidence (p < 0.05) of a negative correlation between Lingcod CPUE and wind stress on the day prior to HnL sampling that was robust to the removal of the potential outlier (Figure 9B, column 1).
There was evidence (p < 0.05) of a negative correlation between Lingcod BPUE and wind stress on the day of (and day prior to) HnL sampling. These significant correlations were robust to the removal of the potential outlier (Figure 9A&B, column 2).
When looking across site-days, there were no significant relationships between fish response variables and wind stress on the day of HnL sampling (figure 10A). However, there was a significant negative correlation between Lingcod CPUE and BPUE and wind stress on the day prior to HnL sampling (Figure 10B, column 1 and 2). The removal of an outlier rendered the BPUE relationship not statistically significant.
The four other focal species – Blue/Deacon Rockfish, China Rockfish, Yelloweye Rockfish and Cabezon – had low catch rates across all sites – and therefore had sample sizes too low to statistically analyze. However, no apparent trends were seen with the data available (see figures 11-14).

Figure 11. Mean CPUE, BPUE, and length by N-S wind stress for all focal species. Each data point represents one site-day. Negative N-S wind stress values (to the left of the vertical line) indicate upwelling.

Figure 12. Mean CPUE, BPUE, and length by one day lagged N-S wind stress for all focal species. Each data point represents one site-day. Negative N-S wind stress values (to the left of the vertical line) indicate upwelling.

Figure 13. Mean CPUE, BPUE, and length by N-S wind stress for all focal species. Each data point represents one day. Negative N-S wind stress values (to the left of the vertical line) indicate upwelling.

Figure 14. Mean CPUE, BPUE, and length by one day lagged N-S wind stress for all focal species. Each data point represents one day. Negative N-S wind stress values (to the left of the vertical line) indicate upwelling.

4.3 Bottom Temperature

Figure 15. Bottom temperature (black) and lagged bottom temperature (blue) for each of the HnL sampling days in 2014, 2015, 2017 & 2019.

4.3.1 Aggregate Data

Data are shown aggregated at two levels:

By day (one value per day, n=26)
By site-day (one value per site per day, n=53)

4.3.1.1 By day

Figure 16. Mean aggregate CPUE and BPUE by bottom temperature and lagged bottom temperature. Each data point represents one day. Spearman’s rank correlation coefficient is reported.

4.3.1.2 By site-day

Figure 17. Mean aggregate CPUE and BPUE by bottom temperature and lagged bottom temperature. Each data point represents one site-day. Spearman’s rank correlation coefficient is reported.

There was no apparent trend in aggregate CPUE or BPUE with bottom temperature values observed at the Cape Falcon Marine Reserve or associated comparison areas.

There were no statistically significant correlations between mean CPUE or BPUE, aggregated to the day or site-day level, and bottom temperature on the day of (or day prior to) HnL sampling.

4.3.2 Focal Species

Mean CPUE, BPUE and size are plotted below as a function of bottom temperature on the day of (and day prior) to HnL sampling for both Black Rockfish and Lingcod.

4.3.2.1 By day

Figure 18. Mean CPUE, BPUE, and length by bottom temperature (A) and lagged bottom temperature (B) for two focal species, Black Rockfish and Lingcod. Each data point represents one day. Spearman’s rank correlation coefficient is reported.

4.3.2.2 By site-day

Figure 19. Mean CPUE, BPUE, and length by bottom temperature (A) and lagged bottom temperature (B) for two focal species, Black Rockfish and Lingcod. Each data point represents one site-day. Spearman’s rank correlation coefficient is reported.

There was no evidence of correlation between Black Rockfish or Lingcod responses and bottom temperature on the day of (or day prior to) HnL sampling.

There were no statistically significant correlations between CPUE, BPUE, or mean size and bottom temperature on the day of (or day prior to) HnL sampling for Lingcod or Black Rockfish.
The four other focal species – Blue/Deacon Rockfish, China Rockfish, Yelloweye Rockfish and Cabezon – had low catch rates across all sites – and therefore had sample sizes too low to statistically analyze. However, no apparent trends were seen with the data available (see figures 20-23).

Figure 20. Mean CPUE, BPUE, and length by bottom temperature for all focal species. Each data point represents one site-day.

Figure 21. Mean CPUE, BPUE, and length by one day lagged bottom temperature for all focal species. Each data point represents one site-day.

Figure 22. Mean CPUE, BPUE, and length by bottom temperature for all focal species. Each data point represents one day.

Figure 23. Mean CPUE, BPUE, and length by one day lagged bottom temperature for all focal species. Each data point represents one day.

5 Discussion

Dissolved Oxygen

We did not find statistically significant correlations between dissolved oxygen and fish responses (CPUE, biomass, mean size) in either of our two abundant focal species (Black Rockfish and Lingcod). We observed a relatively limited range (4.3-6.2ml/L) of dissolved oxygen values at Cape Falcon Marine Reserve and the associated comparison areas; however, we did not observe hypoxic waters. This may explain why we did not see significant fish responses to dissolved oxygen levels.

Wind stress/Upwelling

We found a statistically significant negative correlation between Lagged N-S wind stress and total aggregate BPUE, where more negative values of N-S wind stress (upwelling) were associated with increased BPUE. This pattern was observed in one of our focal species, Lingcod, where more negative values of both wind stress and Lagged wind stress were associated with higher CPUE, BPUE and larger mean size. We did not find statistically significant correlations for Black Rockfish.

Because Black Rockfish are mid-water schooling fish and feed on pelagic invertebrates, we might expect Black Rockfish CPUE and BPUE to be correlated with wind stress. Increases in upwelling-driven productivity could increase feeding rates in Black Rockfish and increase CPUE; however, we lack information about appropriate time lags needed to align wind stress with primary production, secondary production, and Black Rockfish feeding.

It is possible that the correlation we observed between wind stress and Lingcod CPUE, BPUE, and mean size are related to increased feeding rates during periods of upwelling. Upwelling brings deep, nutrient rich waters to the surface that initiate phytoplankton blooms. However, when the upwelled waters first arrive and phytoplankton blooms have not occurred yet, the water is anomalously clear. Because Lingcod are visual predators, the clear water associated with the onset of upwelling may increase feeding rates and thus increase CPUE and BPUE.

Bottom Temperature

The highest bottom temperatures were observed in 2019. This may be a signal of the 2019 Marine Heatwave, which recorded anomalously warm waters and low levels of upper ocean mixing (i.e., the “Blob 2.0”; Amaya et al., 2020). We did not find statistically significant correlations between bottom temperature and fish responses (CPUE, BPUE, mean size) in either Black Rockfish or Lingcod.

5.1 Considerations

Little is known about the oceanographic conditions in and around the Cape Falcon Marine Reserve. The ODFW Marine Reserves Program has documented some of the first nearshore oceanographic information from this area of Oregon state waters. These efforts have allowed us to explore how changing ocean conditions influence fish response as seen through HnL monitoring data. While we detected some evidence of fish response to changing ocean conditions with wind stress (proxy for upwelling), this analysis merely scratched the surface of possibilities to understanding short-term fish responses to changing ocean conditions. However, our results highlight the value and importance of long-term monitoring as our short-term responses help us predict and understand long-term ecosystem changes.

This report documents the use of multiple sources of oceanographic data to explore relationships with fish response. Oceanographic monitoring in the Cape Falcon Marine Reserve began in late 2018, but HnL surveys began in 2014. This mismatch in timing is common in the start-up of monitoring activities. We consulted oceanographic colleagues who found creative ways to provide relevant data from other sources. Leveraging alternative data sources is a useful strategy worth exploring at other marine reserve sites when in situ oceanographic data are missing. The strength of the ODFW Monitoring program lies in biological data collection, and finding ways to incorporate relevant oceanographic datasets can provide valuable insights into changes in the nearshore ocean ecosystem.

In contrast to our findings at Cape Perpetua (Cape Perpetua Hook and Line Oceanography Appendix), we found a correlation of stress (proxy for upwelling) and fish response (at the aggregate level and for Lingcod), but no correlation of fish response with bottom dissolved oxygen or temperature. The significant correlation of wind stress at Cape Falcon and a lack of one at Cape Perpetua is surprising. In both locations there were observed conditions that represent upwelling, but we did not observe similar fish responses. However, with respect to dissolved oxygen, our observations supported our hypothesis that decreased CPUE and BPUE would be observed during periods of hypoxia, and that there would be no significant differences in fish response during periods where no hypoxic conditions were observed. Cape Perpetua is an area known to be affected annually by low oxygen and low pH waters; but little is known about dissolved oxygen conditions in and around Cape Falcon. We observed no instances of hypoxia on (or immediately before) HnL sampling dates at the Cape Falcon Marine Reserve, and found no evidence of fish response, whereas we did observe hypoxic conditions at Cape Perpetua as well as evidence of fish response. Maintaining coupled oceanographic and biological monitoring is critical to increase our understanding about fish responses to hypoxia in economically important marine fishes.

6 Acknowledgements

Thank you to Dr. Jack Barth and Dr. Francis Chan who contributed PISCO data and provided feedback on a draft report.

7 References

Allen, J. S. (1980). Models of wind-driven currents on the continental shelf. Annual Review of Fluid Mechanics 12, 389–433.

Amaya, D. J., Miller, A. J., Xie, S.-P., and Kosaka, Y. (2020). Physical drivers of the summer 2019 North Pacific marine heatwave. Nature communications 11, 1–9.

Bakun, A., Black, B. A., Bograd, S. J., Garcia-Reyes, M., Miller, A. J., Rykaczewski, R. R., & Sydeman, W. J. (2015). Anticipated effects of climate change on coastal upwelling ecosystems. Current Climate Change Reports, 1(2), 85-93.

National Data Buoy Center (1971). Meteorological and oceanographic data collected from the National Data Buoy Center Coastal-Marine Automated Network (C-MAN) and moored (weather) buoys. North-South wind stress at NWPO3 and TLBO3 . NOAA National Centers for Environmental Information. Dataset. https://www.ncei.noaa.gov/archive/accession/NDBC-CMANWx. Accessed 2021.

Rabalais, N. N., Diaz, R. J., Levin, L. A., Turner, R. E., Gilbert, D., and Zhang, J. (2010). Dynamics and distribution of natural and human-caused hypoxia. Biogeosciences 7, 585–619.

Samelson, R., Barbour, P., Barth, J., Bielli, S., Boyd, T., Chelton, D., et al. (2002). Wind stress forcing of the Oregon coastal ocean during the 1999 upwelling season. Journal of Geophysical Research: Oceans 107, 2–1.