This research aims to quantify the abundance of settling temperate reef fishes and identify the community composition to determine the value of designated marine reserve habitats for protecting fishes during their early life stages. Settlement rates of juvenile fishes are known to be highly variable seasonally, among years, and by species. Therefore, long-term settlement data sets are valuable to document natural variation in population replenishment as well as short- to long-term effects of oceanography and fishing pressure, especially for socio-economically important fisheries species where forecasting population abundance is crucial for sustainable management of their stocks. In two pilot years (2011, 2012) at the Otter Rock Marine Reserve site, both immediately before and after reserve designation, Dr. Kirsten Grorud-Colvert (OSU) began a project using Standard Monitoring Units for the Recruitment of Fishes or SMURFs (Ammann 2004) to sample the relative abundance of settlement-stage juvenile fishes in the marine reserve and associated comparison area at Cape Foulweather. This was the first successful use of these sampling tools off the coast of Oregon. In 2013, ODFW became a collaborative partner in this effort, along with the Oregon Coast Aquarium and the lab of Dr. Su Sponaugle (OSU, Hatfield Marine Science Center). In 2014, SMURFing was established at Redfish Rocks to add a second marine reserve to the annual monitoring effort, along with its associated comparison area at Humbug Mountain. Additional collaborations have occurred with Dr. Scott Heppell (OSU, Department of Fisheries and Wildlife) and with Dr. Steve Rumrill (ODFW Shellfish Program Leader) and Dr. Aaron Galloway (OIMB), who have used these SMURF moorings to sample larval invertebrates. Oceanographic sensors were added to the moorings in 2014.
This project has successfully monitored settlement-stage fishes for nine years (2011-2019) and is ongoing. Throughout this time, we have completed 130 days of sampling and have sampled >10,000 fishes. We have developed productive collaborations with other research groups within the ODFW Marine Program and the NMFS Fishery Resource Analysis and Monitoring Division. Our data have contributed to an ODFW Internal Report on cabezon age and growth (Rasmuson et al. 2019) and to the 2019 Cabezon Stock Assessment (Cope et al. 2019). This project has benefitted from local expertise via partnerships we have built within local communities including the Oregon Coast Aquarium, commercial/recreational boat captains in Port Orford, and the OSU Port Orford Field Station. This program has contributed to the training and dissertation research of three graduate students, a number of undergraduate students (4 OSU undergrads, 3 NSF REU’s, 2 Sea Grant Summer Scholars, and students enrolled in an OSU Environmental Science Field Course), and one Port Orford High School student. In addition, it has provided hands-on research experience for local volunteers and community members. Our project and data have been presented to the scientific community via three peer-reviewed publications (Ottmann et al. 2016, 2018, 2019), one M.S. thesis (Ottmann 2018), one PhD dissertation (Fennie 2020), with another in progress (Wilson), and at several conferences (State of the Coast: Ottmann et al. 2015; American Fisheries Society Meeting: Ottmann et al., 2015; Western Groundfish Conference: Ottmann et al. 2016; Western Society of Naturalists: Wilson et al. 2020; 44th Larval Fish Conference: Wilson et al. 2021). We have also shared our project in many public outreach events (e.g., Cape Perpetua Collaborative Young Scientist Seminar, ODFW Slice of Science, HMSC Marine Science Day, Corvallis K-12 Science Night, Finding the Hook: Teaching STEAM Using Oregon’s Sustainable Fisheries).
Diversity and Abundance:
Settlement Patterns:
SMURFs were designed to sample settlement-stage juvenile fishes that are often found in kelp habitats immediately post-settlement. Of the Oregon marine reserves, the Otter Rock Marine Reserve and Redfish Rocks Marine Reserves are the primary targets for this sampling due to the presence of kelp habitats in these reserves. In addition, these central and southern Oregon locations allow for comparison of settlement patterns on either side of Cape Blanco, a known biogeographical break (Checkley and Barth 2009).
SMURF collectors consist of black polyvinyl chloride mesh folded inside a long (100 x 35 cm) cylinder of garden fencing, forming a 3-D structure that simulates natural settlement substrates such as a kelp canopy. These collectors provide an artificial refuge for settlement-size fishes high in the water column.
The research design includes four SMURFs deployed at each reserve and comparison area to balance replication, suitable habitats for settlement and benthic mooring placement, and spacing of ~ 300m between SMURFs (Figure 1). SMURFs are deployed 1m below the surface by attaching them to a mooring anchored in sandy substrates in ~15m of water, 390-1,200m from shore. The deployment locations were selected at a conservative distance offshore of underwater boulders and kelp canopy to ensure direct settlement of fish from the water column to SMURFs as opposed to their movement up to SMURFs from the surrounding substrate.
New fish settlers are collected from the SMURFs every 2 wks during the summer settlement season, which starts in April-June (weather dependent) and extends into September (Love 2011). Species identification is based on external, morphological characteristics (meristics); however, meristics are not always sufficient to identify certain rockfish species as juveniles. Thus, the following species are grouped into three complexes: OYTB, which includes olive, yellowtail, and black rockfishes; QGBCC, which includes quillback, gopher, black-and-yellow, copper, and china rockfishes; and SR, which includes splitnose and redbanded rockfishes. The OYTB and QGBCC complexes reflect similar groupings used in previous research off the coast of California, allowing us to compare results (BYO and KGBC; Wilson et al. 2008, Caselle et al. 2010a). The Oregon complexes add china and quillback rockfishes and do not include kelp rockfishes, as the latter species does not extend beyond central California, the northern limit of their adult distribution (Eschmeyer et al. 1999).
Species presence within these complexes has been verified through genetic identification at the National Oceanic and Atmospheric Administration (NOAA) Southwest Fisheries Science Center in Santa Cruz, California. Additional species collected in high numbers in Oregon SMURFs are cabezon and tiger rockfish.
All analyses were conducted and figures were created with R and R studio (R Core Team, 2020; RStudio Team, 2020).
Research questions were addressed using the top five most abundant species(complexes) (cabezon, OYTB, QGBC, SR, and tiger rockfishes). Settlement rate was calculated using the number of fish per SMURF collector per day, calculated using the total number of days in the sampling period since the last collection. Settlement variability was assessed at two spatial levels: site and region. Within each site, we assessed annual variability. Within each region, two temporal levels of variability are assessed: annual and seasonal. Because settlement is highly variable by species(complex) we discussed spatial and temporal patterns separately for each species(complex). We used data visualizations and statistical comparisons to address spatial and temporal patterns.
Site
Data Visualization: We used paired bar plots with standard error bars to visualize how settlement rate differs by site. To illustrate species(complex)-specific patterns in settlement by site, we included columns for species(complex).
Statistical Comparison: We used a Generalized Linear Mixed Model (GLMM) to statistically compare settlement between sites. We modeled settlement rate using the count of fishes caught with an offset for the sampling interval and number of moorings deployed. We fit separate models for each species/region combination and used a combination of Poisson or negative binomial distributions depending on best fit. Because we were interested in site differences, year and day-of-year were included as random effects to account for variability across years and within the settlement season (sensu Ottmann et al., 2018). Because sampling did not start in southern Oregon until 2014, the southern Oregon site comparison only includes data from 2014-2019, while the central Oregon site comparison includes data from 2012-2019.
Annual patterns
Region
Data Visualization: We used bar plots with standard error bars to visualize how settlement rate differs region. We included colors for annual differences and columns for species(complex) differences.
Statistical Comparison: We used a GLMM to statistically compare settlement rate between regions. We modeled settlement rate using the count of fishes caught with an offset for the sampling interval and number of moorings deployed (as above), with a negative binomial distribution. We fit separate models for each species. Because we were interested in regional differences, year and day-of-year were included as random effects to account for variability across years and within the settlement season (sensu Ottmann et al. 2018). We restricted our analysis to 2014-2019, when both regions were sampled.
Annual patterns
Seasonal patterns
We selected a focal species and region (cabezon collected from the central Oregon region) to explore the relationship between oceanography and settlement rate. We chose this focal species and region because we have the longest data set in the central region, and cabezon are consistently present in moderate numbers each year, maximizing the opportunity for statistical power. Additionally, a PhD student in the SMURF research group (Megan Wilson) is currently studying the oceanographic and ecological factors that affect the early life stages of Cabezon using otolith microstructure analysis.
In situ temperature Settlement rate may be correlated to in situ temperature if distinct bodies of water are favorable for settlement and have a distinct thermal signature. In addition, because fishes are ectotherms, temperature may affect early growth rates and settlement. We calculated the Pearson correlation coefficient between the mean settlement rate of cabezon and average mid-depth temperature. Temperature data were collected using Hobo sensors at mid-depth (~8m below the surface) on several SMURF moorings, and these were averaged across all moorings and sites. We included the average temperature value on the day that sampling occurred, with a 1-, 2-, and 3-day lag. We also included the average temperature over the five days preceding sampling, and over a 15-day interval (corresponding to a sampling interval). To date, there are no published studies that identify thermal signatures of water bodies favorable for cabezon growth and/or settlement (but see Schroeder et al. 2020 for rockfishes). Therefore, to test a range of time scales in which we might see a temperature effect on cabezon settlement, we included lagged and averaged temperature values in our analysis.
Other oceanographic variables Partial Least Squares Regression (PLSR) is a tool used to assess the percent of variance in one or more response variables that is explained by multiple predictor variables. We used PLSR to analyze the relationship between the mean annual settlement rate of central Oregon cabezon and nine oceanographic variables including the Pacific Decadal Oscillation (PDO), Ocean Nino Index (ONI), Nearshore Ichthyoplankton biomass (indicator of overall larval fish abundance), North and South copepod biomass anomalies (indicators of prey conditions), the Biologically Effective Upwelling Index (BEUTI; indicator of productivity), the Coastal Upwelling Transport Index (CUTI; indicator of along-shore and cross-self water transport), Winter Sea Surface Temperature (relevant to larval and early juvenile development), and the Spring Transition Index (STI; indicator of the onset of seasonal productivity) (Table 5). These oceanographic variables were included because they have been hypothesized to affect fish early growth and development and therefore may serve as indicators of settlement success (Wilson et al. 2008; Caselle et al. 2010, Markel and Shurin 2020).
Through this SMURF project, we have documented a diverse community of settlement-stage fishes arriving to Oregon’s nearshore habitat. Over the past 8 years of study, we collected 10,994 fishes from 10 different species or species complexes. Tables 1 -3 below illustrate the diversity of fishes captured by SMURFs and a large degree of variation in abundance by species(complex), year, region, and site. We identified Boccacio rockfish as a rare species in our collections. Boccacio accounted for 1% of the total species composition in the south region in 2015 and 2019 and in the central region in 2017 but were otherwise absent. Additionally, kelp clingfish was identified as a unique species, occurring only in the south region in 2019. We identified the five most abundant species to be cabezon and OYTB, QGBC, tiger, and SR rockfishes.
Take-away: SMURFs are effective tools to sample the community of nearshore settlement-stage fishes. There is variability in abundance by species(complex), year, region, and site.
The abundance of settlement-stage fishes varies by year, region, and site. Looking across species and region, the total number of fishes collected ranges from 391 - 3886 (Table 3.1.1). Within a given species, there is variation in abundance by year, region, and site. For example, focusing on cabezon, abundance ranges between 24 - 440 in the central region and 45 - 606 in the southern region, with the peak abundance occurring in different years in the different regions (Table 3.1.2). For cabezon at Otter Rock Marine Reserve and Cape Foulweather Comparison Area sites, the pattern of interannual variability tracks between sites with only slight differences in abundance (Table 3.1.3). This can be contrasted with SR rockfish, where interannual variability varies between sites: some years had high abundance at Otter Rock Marine Reserve but not at Cape Foulweather Comparison Area (see also section 4.1.1). Within each year, region, and site, each species(complex) has a different pattern of abundance. For example, the abundance of kelp greenling in the central region ranges from 0-16 individuals across all years, while the abundance of QGBC in the central region ranges from 2-668 individuals across all years. Note that the category “unID” refers to individuals that we were unable to identify to species(complex).
Total Fishes | |
---|---|
Central | |
2012 | 391 |
2013 | 1530 |
2014 | 404 |
2015 | 2022 |
2016 | 2720 |
2017 | 536 |
2018 | 592 |
2019 | 1810 |
South | |
2014 | 906 |
2015 | 3674 |
2016 | 3886 |
2017 | 1212 |
2018 | 572 |
2019 | 1732 |
Year
|
|||||||||
---|---|---|---|---|---|---|---|---|---|
2012 | 2013 | 2014 | 2015 | 2016 | 2017 | 2018 | 2019 | Species Total | |
Central | |||||||||
Bocaccio | 0 | 0 | 0 | 1 | 1 | 2 | 0 | 0 | 4 |
Cabezon | 24 | 127 | 65 | 241 | 267 | 46 | 111 | 440 | 1321 |
Kelp Clingfish | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Kelp Greenling | 0 | 3 | 0 | 1 | 2 | 11 | 16 | 13 | 46 |
Northern Clingfish | 0 | 0 | 0 | 0 | 0 | 76 | 3 | 0 | 79 |
OYTB | 1 | 16 | 125 | 13 | 279 | 90 | 99 | 220 | 843 |
QGBC | 116 | 52 | 2 | 203 | 668 | 9 | 53 | 218 | 1321 |
Snailfish | 6 | 1 | 1 | 0 | 1 | 33 | 8 | 1 | 51 |
SR | 32 | 552 | 1 | 362 | 37 | 0 | 4 | 2 | 990 |
Tiger | 11 | 10 | 6 | 188 | 97 | 1 | 2 | 11 | 326 |
unID | 5 | 4 | 2 | 2 | 8 | 0 | 0 | 0 | 21 |
Total | 196 | 765 | 202 | 1011 | 1360 | 268 | 296 | 905 | 5003 |
South | |||||||||
Bocaccio | NA | NA | 0 | 18 | 1 | 1 | 0 | 11 | 31 |
Cabezon | NA | NA | 232 | 606 | 220 | 45 | 139 | 219 | 1461 |
Kelp Clingfish | NA | NA | 0 | 0 | 0 | 0 | 0 | 42 | 42 |
Kelp Greenling | NA | NA | 10 | 27 | 23 | 10 | 6 | 21 | 97 |
Northern Clingfish | NA | NA | 0 | 0 | 0 | 2 | 22 | 38 | 62 |
OYTB | NA | NA | 191 | 16 | 722 | 528 | 93 | 341 | 1891 |
QGBC | NA | NA | 18 | 256 | 305 | 16 | 13 | 190 | 798 |
Snailfish | NA | NA | 1 | 6 | 1 | 1 | 12 | 1 | 22 |
SR | NA | NA | 0 | 878 | 542 | 1 | 0 | 3 | 1424 |
Tiger | NA | NA | 0 | 27 | 110 | 0 | 0 | 0 | 137 |
unID | NA | NA | 1 | 3 | 19 | 2 | 1 | 0 | 26 |
Total | NA | NA | 453 | 1837 | 1943 | 606 | 286 | 866 | 5991 |
Year
|
|||||||||
---|---|---|---|---|---|---|---|---|---|
2012 | 2013 | 2014 | 2015 | 2016 | 2017 | 2018 | 2019 | Species Total | |
Cape Foulweather Comparison Area | |||||||||
Bocaccio | 0 | 0 | 0 | 0 | 1 | 2 | 0 | 0 | 3 |
Cabezon | 10 | 77 | 27 | 122 | 137 | 26 | 59 | 233 | 691 |
Kelp Clingfish | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Kelp Greenling | 0 | 2 | 0 | 1 | 1 | 5 | 6 | 7 | 22 |
Northern Clingfish | 0 | 0 | 0 | 0 | 0 | 10 | 1 | 0 | 11 |
OYTB | 0 | 4 | 58 | 5 | 121 | 26 | 28 | 129 | 371 |
QGBC | 75 | 27 | 0 | 106 | 369 | 4 | 23 | 101 | 705 |
Snailfish | 0 | 1 | 0 | 0 | 1 | 5 | 2 | 0 | 9 |
SR | 9 | 40 | 0 | 54 | 35 | 0 | 0 | 1 | 139 |
Tiger | 1 | 4 | 5 | 57 | 70 | 1 | 1 | 4 | 143 |
unID | 1 | 1 | 1 | 1 | 4 | 0 | 0 | 0 | 8 |
Total | 96 | 156 | 91 | 346 | 739 | 79 | 120 | 475 | 2102 |
Otter Rock Marine Reserve | |||||||||
Bocaccio | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 1 |
Cabezon | 14 | 50 | 38 | 119 | 130 | 20 | 52 | 207 | 630 |
Kelp Clingfish | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Kelp Greenling | 0 | 1 | 0 | 0 | 1 | 6 | 10 | 6 | 24 |
Northern Clingfish | 0 | 0 | 0 | 0 | 0 | 66 | 2 | 0 | 68 |
OYTB | 1 | 12 | 67 | 8 | 158 | 64 | 71 | 91 | 472 |
QGBC | 41 | 25 | 2 | 97 | 299 | 5 | 30 | 117 | 616 |
Snailfish | 6 | 0 | 1 | 0 | 0 | 28 | 6 | 1 | 42 |
SR | 23 | 512 | 1 | 308 | 2 | 0 | 4 | 1 | 851 |
Tiger | 10 | 6 | 1 | 131 | 27 | 0 | 1 | 7 | 183 |
unID | 4 | 3 | 1 | 1 | 4 | 0 | 0 | 0 | 13 |
Total | 100 | 609 | 111 | 665 | 621 | 189 | 176 | 430 | 2901 |
Humbug Mountain Comparison Area | |||||||||
Bocaccio | NA | NA | 0 | 12 | 0 | 1 | 0 | 9 | 22 |
Cabezon | NA | NA | 139 | 401 | 151 | 27 | 102 | 152 | 972 |
Kelp Clingfish | NA | NA | 0 | 0 | 0 | 0 | 0 | 11 | 11 |
Kelp Greenling | NA | NA | 6 | 13 | 14 | 8 | 3 | 14 | 58 |
Northern Clingfish | NA | NA | 0 | 0 | 0 | 1 | 2 | 25 | 28 |
OYTB | NA | NA | 88 | 6 | 310 | 231 | 39 | 216 | 890 |
QGBC | NA | NA | 15 | 133 | 148 | 0 | 9 | 116 | 421 |
Snailfish | NA | NA | 0 | 0 | 0 | 1 | 7 | 0 | 8 |
SR | NA | NA | 0 | 595 | 203 | 0 | 0 | 0 | 798 |
Tiger | NA | NA | 0 | 20 | 33 | 0 | 0 | 0 | 53 |
unID | NA | NA | 1 | 0 | 17 | 1 | 0 | 0 | 19 |
Total | NA | NA | 249 | 1180 | 876 | 270 | 162 | 543 | 3280 |
Redfish Rocks Marine Reserve | |||||||||
Bocaccio | NA | NA | 0 | 6 | 1 | 0 | 0 | 2 | 9 |
Cabezon | NA | NA | 93 | 205 | 69 | 18 | 37 | 67 | 489 |
Kelp Clingfish | NA | NA | 0 | 0 | 0 | 0 | 0 | 31 | 31 |
Kelp Greenling | NA | NA | 4 | 14 | 9 | 2 | 3 | 7 | 39 |
Northern Clingfish | NA | NA | 0 | 0 | 0 | 1 | 20 | 13 | 34 |
OYTB | NA | NA | 103 | 10 | 412 | 297 | 54 | 125 | 1001 |
QGBC | NA | NA | 3 | 123 | 157 | 16 | 4 | 74 | 377 |
Snailfish | NA | NA | 1 | 6 | 1 | 0 | 5 | 1 | 14 |
SR | NA | NA | 0 | 283 | 339 | 1 | 0 | 3 | 626 |
Tiger | NA | NA | 0 | 7 | 77 | 0 | 0 | 0 | 84 |
unID | NA | NA | 0 | 3 | 2 | 1 | 1 | 0 | 7 |
Total | NA | NA | 204 | 657 | 1067 | 336 | 124 | 323 | 2711 |
Take-away: The species(complex) composition of nearshore settlement-stages fishes arriving to the Oregon nearshore is dynamic, changing substantially by year, region, and site.
The species composition of settlement-stage fishes varies by year, region, and site. Only species(complexes) that made up > 1% of the annual catch were included. The number of species(complexes) present each year is variable, ranging from 4 (e.g., south 2014) to 8 (e.g., central 2018). It is important to note that these figures do not reflect the total magnitude of settlement in each region or year – i.e., a greater number of species(complexes) present does not necessarily indicate higher settlement rates. For example, 8 species(complexes) and 592 total fishes were collected in central Oregon in 2018 whereas 6 species(complexes) and 2720 total fishes were collected in central Oregon in 2016.
Take-away: There are significant differences in settlement rate between sites for certain species(complexes). Site-specific patterns of settlement are variable across years.
For OYTB and SR rockfishes, settlement was variable but greater overall in Otter Rock Marine Reserve compared to Cape Foulweather Comparison Area. For tiger rockfish, settlement was variable but greater overall in Redfish Rocks Marine Reserve compared to Humbug Mountain Comparison Area. For cabezon, settlement was variable but greater overall in Humbug Mountain Comparison Area compared to Redfish Rocks Marine Reserve.
OYTB settlement rate was 1.25 times higher in Otter Rock Marine Reserve compared to Cape Foulweather Comparison Area (GLMM; p < 0.01), and SR settlement rate was 5.13 times higher in Otter Rock Marine Reserve compared to Cape Foulweather Comparison Area (GLMM; p < 0.001). There were no significant differences detected for the other species(complexes).
Cabezon settlement rate was lower in Redfish Rocks Marine Reserve by a factor of 0.52 (GLMM; p < 0.001), and tiger rockfish settlement rate was 1.99 times higher in Redfish Rocks Marine Reserve than in Humbug Mountain Comparison Area (GLMM; p < 0.001). There were no significant differences detected for the other species(complexes).
Most species(complexes) followed similar annual patterns at both Otter Rock Marine Reserve and Cape Foulweather Comparison Area. Notably, SR rockfishes at Otter Rock Marine Reserve had a much higher settlement rate than at Cape Foulweather Comparison Area in 2013 and 2015.
QGBC, OYTB, and tiger rockfishes followed similar annual patterns at Redfish Rocks Marine Reserve and Humbug Mountain Comparison Area. Cabezon had higher abundance in Humbug Mountain Comparison Area for all years except 2015, and SR rockfishes had slightly higher abundance in Humbug Mountain Comparison Area in 2015.
Take-away: There are significant differences in settlement rate between central and southern Oregon for certain fish species(complexes).
Considering all years combined, there were significant differences in settlement rate between the central and southern regions in OYTB and Tiger rockfish. The settlement rate of OYTB was 5.77 times higher in southern Oregon than in central Oregon; the settlement rate of QGBC rockfish was 2.08 times higher in central Oregon, and the settlement rate of tiger rockfish was lower in southern Oregon by a factor of 0.07 (GLMM; p < 0.05). There were no significant differences by region in the other species(complexes) (GLMM; p > 0.05). For this analysis we truncated our data set to only include years when both regions were sampled (2014-2019).
Take-away: Settlement patterns are highly variable among years and regions, illustrating fluctuation in population replenishment rates in space and time for different species(complexes).
Settlement patterns were highly variable each year, and this variability was species(complex)-dependent. Cabezon were present in each year of sampling, with a relatively low degree of variation (0.04 - 0.47 fish/SMURF/day). OYTB and QGBC rockfishes were present in nearly every year of sampling but had a relatively higher degree of interannual variation (e.g., OYTB: 0 - 0.71 fish/SMURF/day). Finally, SR and tiger rockfish were only present in some years and had a high degree of variation across years (e.g., SR: 0 - 0.43 fish/SMURF/day). In addition, the highest settlement years were not the same across the species(complexes), suggesting taxon-specific (e.g., life history based) responses to environmental conditions. Interestingly, the maximum settlement rate was approximately consistent across taxa (~ 0.40 fish/SMURF/day) with the exception of OYTB, which settled at a rate of 0.7 fish/SMURF/day in 2016 at our southern Oregon sites.
We found broad annual trends in the settlement rate of cabezon, OYTB, and QGBC (Figure 11). Southern region cabezon had the highest settlement in 2014, with settlement lower thereafter. Central region cabezon had the opposite pattern: low settlement in 2014 with a gradual increase through time. Southern region OYTB had a very high peak in settlement rate in 2016, with lower settlement before and after. Central region OYTB had a similar, but much smaller, peak in 2016. QGBC rockfish in the south had a similar peak in 2016 with lower settlement in recent years, but in contrast to OYTB and southern QGBC, settlement was high in central QGBC before the peak (Figure 12). Within these broad trends, there is a large degree of variability across years, as noted above. To investigate specific annual contrasts by species(complex), a full list of significant differences in annual pairwise contrasts is presented in Table 4.
Region | Contrast | Cabezon | OYTB | QGBC | SR | Tiger |
---|---|---|---|---|---|---|
Central | 2013 / 2012 | ns | ns | ns | ns | ns |
2014 / 2012 | ns | ns | ns | ns | ns | |
2014 / 2013 |
|
ns | ns | ns | ns | |
2015 / 2012 | ns | ns | ns | ns | ns | |
2015 / 2013 | ns | ns | *** | *** | *** | |
2015 / 2014 | ns | ns | ** | ns | ns | |
2016 / 2012 | ns | ns | ns | ns | ns | |
2016 / 2013 | ns | *** | ** |
|
ns | |
2016 / 2014 | ** | *** | ns | ns | ns | |
2016 / 2015 |
|
*** |
|
ns | ns | |
2017 / 2012 | ns | ns | ns | ns | ns | |
2017 / 2013 | *** | ns | ns | ns | ns | |
2017 / 2014 | ns | ns | ns | ns | ns | |
2017 / 2015 | *** | ** | *** | ns |
|
|
2017 / 2016 | *** | *** | ** | ns | ns | |
2018 / 2012 | ns | ns | ns | ns | ns | |
2018 / 2013 | ** | ns | ns | ns | ns | |
2018 / 2014 | ns | ns | ns | ns | ns | |
2018 / 2015 | ns | ** |
|
ns | ns | |
2018 / 2016 | ** | *** | ns | ns | ns | |
2018 / 2017 | ns | ns | ** | ns | ns | |
2019 / 2012 | ** | ns | ns | ns | ns | |
2019 / 2013 | ** | *** | ** | ns | ns | |
2019 / 2014 | *** | *** |
|
ns | ns | |
2019 / 2015 | *** | *** | ns | ns | ns | |
2019 / 2016 | ** | ns | ns | ns | ns | |
2019 / 2017 | *** | *** | *** | ns | ns | |
2019 / 2018 | *** | *** | ns | ns | ns | |
South | 2015 / 2014 | ns | ns | *** | ||
2016 / 2014 | ** | ns | ** | |||
2016 / 2015 | *** | ns | ** | |||
2017 / 2014 | ** | ns | ns | |||
2017 / 2015 | *** | ns | ** | |||
2017 / 2016 | ns | ns | ns | |||
2018 / 2014 | ns | ns | ns | |||
2018 / 2015 |
|
|
** | |||
2018 / 2016 | ns | ns | ns | |||
2018 / 2017 | ns | ns | ns | |||
2019 / 2014 | ns | ns | *** | |||
2019 / 2015 | ns | ns | ns | |||
2019 / 2016 | ** | ns | ** | |||
2019 / 2017 | ** | ns | ** | |||
2019 / 2018 | ns | ns | ** |
GAMs did not appropriately converge and failed to capture variation in SR and tiger rockfishes, possibly because of the high degree of variability between years and the large number zeros in the settlement data for these species(complexes). Therefore, for SR and tiger rockfishes, we treated year as a categorical variable and used GLM to test for significant differences in settlement rate between years. For central region tiger rockfish, there was a settlement peak in 2015, and exceptionally low settlement in 2017. For central region SR rockfish, there was a settlement peak in 2013 and 2015. For southern region tiger rockfish, settlement peaked in 2016 and in southern region SR rockfish, settlement peaked in 2015 and 2016 (Figure 13).
Take-away: Settlement is seasonal, with different settlement-stage fish species(complexes) arriving to nearshore waters at different times in the season.
Settlement is seasonal, with different species and species complexes arriving to nearshore waters at different times throughout the settlement season (April-September). For some groups, a peak in settlement occurred at approximately the same time each year. For example, the highest settlement of OYTB occurred between May and early June, with very low numbers settling afterwards. Settlement of QGBC peaked between June and August for most years but there were additional early and late-season settlement pulses in 2016. Interestingly, cabezon departs from this “single pulse” pattern and settles at a relatively high rate throughout the settlement season, with peaks between late April and early June. Finally, SR and tiger rockfishes are not consistently abundant each year, but in years when they are abundant, these species tend to settle later in the settlement season, starting in July and increasing towards September, possibly peaking in settlement beyond the sampling season.
We selected a focal fish species and region (cabezon collected from the central Oregon region) to explore the relationship between oceanography and settlement rate.
Take-away: For cabezon in the central region, there were no significant correlations between mean settlement rate and mid-depth temperature. An approach that includes multiple oceanographic variables related to early cabezon growth and survival may be more useful in predicting settlement rate.
SMURF Hobo sensors capture daily and interannual variability in water temperature around the SMURF mooring at approximately 8m below the surface.
There was no significant correlation between the settlement rate of cabezon in the central region and mid-depth temperature recorded by the moorings at any of the time lags and/or averages that we calculated.
We used PLSR to analyze the relationship between the mean annual settlement rate of central Cabezon and nine oceanographic variables (Table 5) that are hypothesized to influence cabezon early growth and survival, and therefore impact settlement rate (see Methods).
Variable | Description | Source |
---|---|---|
PDO Winter | Average value of the Pacific Decadal Oscillation (Nov - Feb) | NOAA: California Current Integrated Ecosystem Assessment |
ONI Winter | Average value of the Ocean Nino Index (Nov - Feb) | NOAA: California Current Integrated Ecosystem Assessment |
Nearshore Ichthyo | Nearshore ichthyplankton concentration from Newport Hydrographic line surveys conducted semi-annual to quarterly in Jan-Mar; 0.6m double oblique tow with 333µm mesh size | NOAA: Ocean Ecosystem Indicators of Pacific Salmon Marine Survival in the Northern California Current |
S. Copepod | Southern copepod biomass anomaly | NOAA: California Current Integrated Ecosystem Assessment |
N. Copepod | Northern copepod biomass anomaly | NOAA: California Current Integrated Ecosystem Assessment |
BEUTI | Biologically Effective Upwelling Index | NOAA: California Current Integrated Ecosystem Assessment |
CUTI | Coastal Upwelling Transport Index | NOAA: California Current Integrated Ecosystem Assessment |
SST Winter | Sea Surface Temperature at Stonewall Bank (Nov - Feb) | NOAA: California Current Integrated Ecosystem Assessment |
STI | Annual spring transition index (day of year) | NOAA: California Current Integrated Ecosystem Assessment |
We used three PLSR components to describe the variability in the nine oceanographic variables. ONI winter and BEUTI were top contributors to component 1, PDO winter and CUTI were top contributors to component 2, and ONI winter and PDO winter were top contributors to component 3 (component weights, Table 6). This is illustrated in Figure 18, where settlement rate (Y) is closely aligned with BEUTI and CUTI, which were top contributors to the first and second PLSR components.
Over half of the variability in PDO winter, S. Copepod, BEUTI, and SST winter were captured by the first component. A large degree of variability in N. Copepod, CUTI, and STI were captured by the second component, and variability in Nearshore Ichthyoplankton was captured by the third component (Table 7). The three components together explained 83.8% of the variability in the response variable, mean annual Cabezon settlement rate (last row, Table 7). The regression coefficient (Table 7) indicates the magnitude and direction of the effect of predictor (oceanographic) variables on the response variable (mean annual Cabezon settlement rate).
Variable | PLS Regression Coefficients | C1 Weight2 | C2 Weight2 | C3 Weight2 |
---|---|---|---|---|
PDO Winter | -0.315 | 0.042 | 0.219 | 0.325 |
ONI Winter | 0.599 | 0.183 | 0.161 | 0.624 |
Nearshore Ichthyoplankton | -0.428 | 0.122 | 0.058 | 0.321 |
S. Copepod | -0.021 | 0.049 | 0.081 | 0.000 |
N. Copepod | 0.125 | 0.019 | 0.174 | 0.004 |
BEUTI | 0.086 | 0.183 | 0.019 | 0.083 |
CUTI | 0.425 | 0.158 | 0.364 | 0.036 |
SST Winter | 0.071 | 0.115 | 0.018 | 0.000 |
STI | -0.054 | 0.128 | 0.032 | 0.119 |
Variable | Component.1 | Component.2 | Component.3 |
---|---|---|---|
PDO Winter | 0.667 | 0.870 | 0.871 |
ONI Winter | 0.484 | 0.499 | 0.887 |
Nearshore Ichthyoplankton | 0.401 | 0.438 | 0.698 |
S. Copepod | 0.521 | 0.891 | 0.891 |
N. Copepod | 0.387 | 0.943 | 0.960 |
BEUTI | 0.793 | 0.879 | 0.945 |
CUTI | 0.229 | 0.779 | 0.921 |
SST Winter | 0.776 | 0.926 | 0.937 |
STI | 0.486 | 0.685 | 0.893 |
Settlement Rate | 0.515 | 0.700 | 0.838 |
The SMURF program is a collaborative partnership that has generated a 10-yr (and ongoing) timeseries describing patterns of settlement in several nearshore groundfish species. The SMURF program has facilitated partnerships and knowledge-sharing between ODFW, OSU, and local community members, has resulted in three peer-reviewed publications, one M.S. thesis, one PhD dissertation, with another in progress, and a number of presentations and outreach events. In addition to these graduate students, the program has provided research training for multiple undergraduate students, interns, and volunteers. SMURF sampling effectively captures patterns of abundance and diversity of settlement-stage fishes arriving to Oregon’s nearshore habitats. Throughout our sampling program, we have collected >10,000 fishes and have documented annual, regional (central or southern Oregon), and site-specific (Otter Rock Marine Reserve, Cape Foulweather Comparison Area, Redfish Rocks Marine Reserve, Humbug Mountain Comparison Area) variability in abundance and species(complex) composition of settlement-stage fishes. We have documented significant differences in settlement rate across spatial (regional and site) and temporal (annual and seasonal) scales, and we have explored how settlement patterns vary with changes in oceanographic conditions:
Site: We documented small-scale spatial differences in fish settlement rate for certain species. For some species (OYTB and SR rockfishes), settlement was greater overall in Otter Rock Marine Reserve relative to Cape Foulweather Comparison Area. For one species (tiger rockfish), settlement was greater overall in Redfish Rocks Marine Reserve relative to Humbug Mountain Comparison Area, and for one species (cabezon), settlement was higher in Humbug Mountain Comparison Area. This may be related to habitat differences among sites (e.g., rock, kelp cover, local currents) and/or differences in preferred settlement habitat for these settlement-stage juvenile fishes.
Site: annual patterns: The site-specific patterns of settlement that we observed were variable interannually. In some species, settlement rate appears to vary synchronously at adjacent sites.
Region: Because we sampled across a biogeographic break (Cape Blanco) with known differences in local oceanography, and because studies of several marine taxa have found correlations between settlement patterns and local oceanography, our observed differences in settlement rate by region for tiger, OYTB, and QGBC rockfishes were not surprising. Cabezon and SR did not differ significantly by region, suggesting that other oceanographic, ecological, or biological factors may be influencing their settlement patterns (e.g., location of spawning populations, multiple spawning events, larval and pelagic juvenile dispersal behavior).
Region: annual patterns: Within a region, settlement rate for each species(complex) varied each year. Years with high settlement for some species(complexes) corresponded to low settlement years for other species(complexes). This illustrates interannual fluctuation in population replenishment for each species(complex).
Region: seasonal patterns: Settlement is seasonal, with different species(complexes) arriving to nearshore waters at different times in the season. Some species (e.g., tiger rockfish) settle in a single pulse during the settlement season, and some species arrive in multiple pulses throughout the settlement season (e.g., cabezon).
Oceanographic conditions: The SMURF timeseries is sufficiently long to capture important variability in oceanographic conditions, demonstrating the value of this ongoing timeseries for evaluating the dynamics of ocean change. Different oceanographic variables, or combinations of variables, were important for explaining settlement patterns in central region cabezon and will likely be important for other species(complexes) as well. This is an important area for future research to investigate the emerging effects of a changing ocean (e.g., elevated temperature, increased acidity, decreased dissolved oxygen) on the early life stages and settlement of ecologically and socio-economically important fisheries species.
Additionally, the SMURF settlement timeseries spans before and after reserve implementation, capturing important baseline information about Otter Rock Marine Reserve and Redfish Rocks Marine Reserve. Information about settlement rate before and after reserve implementation is critical to separate out the effects of protection vs. settlement variability. Additionally, variability in settlement, and therefore population replenishment is an important predictor of adult population dynamics. It will be important to contextualize the fluctuations observed in adult populations with the variability in settlement by species (complex) and by year, which we have documented here. It can also assist in teasing apart different effects on the adult population resulting from oceanography, fishing pressure, and protection.
Ammann, A. J. (2004). SMURFs: standard monitoring units for the recruitment of temperate reef fishes. Journal of Experimental Marine Biology and Ecology 299, 135–154.
Caselle, J. E., Carr, M. H., Malone, D. P., Wilson, J. R., and Wendt, D. E. (2010). Can we predict interannual and regional variation in delivery of pelagic juveniles to nearshore populations of rockfishes (genus Sebastes) using simple proxies of ocean conditions? CalCOFI Reports 51 91 - 105.
Checkley Jr, D. M., and Barth, J. A. (2009). Patterns and processes in the California Current System. Progress in Oceanography 83, 49–64.
Cope JM, Berger AM, Whitman AD, Budrick J, Bosely KM, Tsou T, Niles CB, Privitera-Johnson K, Hillier LK, Hinton KE, Wilson MN. 2019. Assessing cabezon (Scorpaenichthys marmoratus) stocks in waters off of California and Oregon, with catch estimation for Washington State. Pacific Fisheries Management Council.
Eschmeyer, W. N., and Herald, E. S. (1999). A field guide to Pacific coast fishes: North America. Houghton Mifflin Harcourt.
Love, M. S. (2011). Certainly more than you want to know about the fishes of the Pacific Coast: a postmodern experience. Really Big Press.
Markel, R. W., and Shurin, J. B. (2020). Contrasting effects of coastal upwelling on growth and recruitment of nearshore Pacific rockfishes (genus Sebastes). Canadian Journal of Fisheries and Aquatic Sciences 77, 950–962.
Ottmann, D., Grorud-Colvert, K., Huntington, B., and Sponaugle, S. (2018). Interannual and regional variability in settlement of groundfishes to protected and fished nearshore waters of Oregon, USA. Marine Ecology Progress Series 598, 131–145.
Ottmann, D., Grorud-Colvert, K., Sard, N. M., Huntington, B. E., Banks, M. A., and Sponaugle, S. (2016). Long-term aggregation of larval fish siblings during dispersal along an open coast. Proceedings of the National Academy of Sciences 113, 14067–14072.
Ottmann, D., Grorud-Colvert, K., and Sponaugle, S. (2019). Age and growth of recently settled splitnose and redbanded rockfishes in the northern California Current. Journal of Sea Research 148, 8–11.
Rasmuson LK, Kautzi L, Aylesworth L, Wilson MN, Grorud-Colvert K. 2019. Age reading of Cabezon (Scorpaenichthys marmoratus): 1) comparison of thin-section and break-and-burn methods and 2) comparison of growth curve fits. ODFW Internal Report.
Schroeder, I. D., Santora, J. A., Bograd, S. J., Hazen, E. L., Sakuma, K. M., Moore, A. M., et al. (2019). Source water variability as a driver of rockfish recruitment in the California Current Ecosystem: implications for climate change and fisheries management. Canadian Journal of Fisheries and Aquatic Sciences 76, 950–960.
Wilson, J. R., Broitman, B. R., Caselle, J. E., and Wendt, D. E. (2008). Recruitment of coastal fishes and oceanographic variability in central California. Estuarine, Coastal and Shelf Science 79, 483–490. doi:10.1016/j.ecss.2008.05.001.