1 Summary: Discontinued Sampling Tools

The ODFW Marine Reserves Program’s ecological monitoring is the first ecosystem-focused, long-term monitoring program to be conducted in Oregon’s nearshore marine environment. In order to accomplish our monitoring goals, we built upon advances in sampling technology and gear to design robust and contemporary survey tools that can function in Oregon’s challenging nearshore environment, and we have research collaborations with scientists who have expertise and contribute to our long-term monitoring efforts. During the development of this long-term monitoring program, we have made adaptations to our data collection based on tool and methods testing, lesson learned in the field and during data analyses, and advice from other scientific experts. In this section, we summarize the use of three discontinued sampling tools:

  • benthic extraction
  • video sled and
  • otolith sampling.

For each tool, we highlight our research goals, methods, and lessons learned. We also document why each was ultimately not incorporated as a core ODFW monitoring tool.

2 Takeaways

Each of our three discontinued tools identified interesting biological patterns that may not have been captured otherwise.

Three initial monitoring tools were ultimately discontinued – benthic extraction, the video sled and otolith sampling, however all three identified new knowledge that may not have been captured otherwise. Using the benthic extraction tool, we were able to document differences in sponge and macroalgae abundance and community composition between areas and in some cases, between site pairs (e.g., Otter Rock Marine Reserve and Cape Foulweather Comparison Area). In addition, our work with the benthic extraction tool documented the occurrence of three species of brown macroalgae that have never before been reported in Oregon waters. Though the taxonomic resolution of the video sled tool was low, we were able to confirm that the abundance and diversity of fish and invertebrates was greatest near rugose rocky habitat. We documented an increase in invertebrate abundance with depth at Otter Rock Marine Reserve and Moolack Comparison Area. We also documented a significantly higher abundance of Sand dollars within Moolack Comparison Area while sea stars and brittle stars were significantly more abundance Otter Rock Marine Reserve, and we documented a significantly higher abundance of sea pens in Redfish Rocks Marine Protected Area compared to Humbug Comparison Area. Finally, our otolith sampling revealed that at all locations, female Black Rockfish tend to reach slightly larger sizes at a given age than males, though this point of divergence occurs at slightly different ages at different locations.

The development of each tool, including baseline data collection and lessons learned in the field and during data analysis, enables the Marine Reserves program to easily reinstate the tools for future research questions.

Each tool was utilized for at least two years, providing the opportunity to learn important lessons about configuration, methodology, external collaboration, and/or data analysis associated with each tool. Though it was ultimately decided that each tool was not conducive to regular ecological monitoring, any of these tools could be reinstated as necessary for future, more specific research questions since our preliminary studies have generated significant institutional knowledge about each method.

We discontinued use of each of these three tools because the resolution of the biological data generated did not meet ecological monitoring goals and/or was not sufficient to justify the operational costs or time investment.

Benthic extraction proved to be a costly endeavor, in terms of sampling equipment, ship-time, and personnel, that ultimately was not conducive to a regular monitoring schedule. While working closely with taxonomic experts proved to be a worthwhile collaboration, the identification of many groups to relevant taxonomic levels (e.g., species) is extremely time intensive and sometimes impossible without genetic confirmation. The video sled was costly to operate and had very poor taxonomic resolution. The otolith sample sizes necessary to document differences in age distribution or age-length relationships between sites were prohibitively large.

3 Benthic Extraction

3.1 Introduction

A benthic biodiversity study in subtidal hard-bottom habitats was conducted to sample the species diversity and abundance of macroinvertebrates and macroalgae not readily captured by our visual survey methods. This sampling approach allows us to resolve species-specific taxonomy for both the algal community and sponge community through collaborative partnership with Dr. Gayle Hansen, a phycologist with Oregon State University, Dr. David Elvin, a sponge taxonomist and lead of the Oregon Porifera Project, and Dr. Nathan Kirk, a researcher at Oregon State University.

We asked the following questions using this benthic extraction data set:

Q1: Does total macroinvertebrate and macroalgal abundance and/or biomass differ between marine reserves and comparison areas?

Q2: Does total macroinvertebrate and macroalgal diversity differ between marine reserves and comparison areas?

Q3: Does community composition of macroinvertebrates and macroalgae differ between marine reserves and comparison areas?

Q4: What species define the macroinvertebrate and macroalgal communities in marine reserves and comparison areas?

Lastly, we were interested in identifying any new records of species occurrence in subtidal nearshore habitats for the state of Oregon.

3.2 Methods

3.2.1 Study Design

In 2011, we conducted surveys for algae, sponges, and other invertebrates at two pairs of co-located sites: 1. Otter Rock Marine Reserve and Cape Foulweather Comparison Area, and 2. Redfish Rocks Marine Reserve and Humbug Mountain Comparison Area. We contracted with local urchin divers to conduct a second round of surveys in 2013, focusing on algae and sponges at Cascade Head Marine Reserve and Cavalier Comparison Area.

We used a stratified random sampling design, restricting the placement of random points to areas of consolidated rock substrate targeting 10m and 15m depths. Ten random points (separated by a minimum of 40m) were generated in both the marine reserve and comparison area sites. In situ transect sampling was then initiated at 6 of these randomly selected points within each site; 3 transects at approximately 10m and 3 transects at approximately 15m depth. If no hard substrate was encountered at the point, the captain maneuvered the boat to the nearest area of consolidated substrate at the appropriate depth (10m or 15m). Three transects were completed per sampling day; a total of 4 days were needed to complete the full set of 12 transects, 6 transects in the marine reserve and 6 transects from the comparison area.

3.2.2 Data Analysis

Mean macroalgal species richness, total biomass (g), and community composition were calculated per transect from replicate quadrat subsamples. Similarly, sponge species richness, total biomass (g), and community composition were calculated per transect from replicate quadrat subsamples. Response variables were transformed if needed to improve normality and homogeneity of variance. Univariate comparisons (t-test) between the reserve and comparison area were conducted on species richness and total biomass (g) response variables for both macroalgae and sponges (Q1 and Q2). The lowest level of taxonomic identification was the species level for macroalgae, family (2011) or genus (2013) for sponges, and phylum for other invertebrates.

To assess differences in community composition between the reserve and comparison area, Bray-Curtis similarity was calculated at the transect scale on both species presence/absence and biomass data. ANOSIM (analysis of similarity), a multivariate analogue of univariate ANOVA, was then conducted on the Bray-Curtis values using PRIMER (v. 6.0) software (Q3). SIMPER was then used to identify which species were primarily responsible for any observed differences between the reserve and comparison area (Q4).

3.3 Results

3.3.1 Sponges

Q1. The biomass of sponges did not differ between any of the site pairs sampled.

  • The total volume of sponges did not differ between Otter Rock Marine Reserve (mean = 444.58 cm³/m² ±278.30 SE) and Cape Foulweather Comparison Area (mean = 433.19 cm³/m² ± 164.38 SE; T-test, t ratio= 0.04, df = 5.09, P = 0.97).

  • The total volume of sponges did not differ between Redfish Rocks Marine Reserve (mean = 62.86 cm3/m2 ± 40.98 cm3/m2 SE) and Humbug Comparison Area (mean = 36.30 cm3/m2 ± 18.60 cm3/m2 SE; T-test, t ratio= 0.59, df = 6.97, P = 0.57).

  • The total volume of sponges did not differ between Cascade Head Comparison Area (mean = 168.4 cm3/m2 ± 74.3 cm3/m2 SE) and Cavalier Comparison Area (mean = 140.7 cm3/m2 ± 54.1 cm3/m2 SE; T-test, t ratio= -0.30, df = 9.14, P = 0.77).

Q2. Average sponge diversity did not differ between any of the site pairs sampled.

  • The diversity of the sponge community did not differ between the Otter Rocks Marine Reserve (mean = 6.67 species/m² ±1.44 SE) and the Foulweather Comparison Area (mean = 12.44 species/m² ±2.08 SE; T-test assuming unequal variance, t ratio= -2.28, df = 7.91, P = 0.0524).

  • The diversity of the sponge community did not differ between the Redfish Rocks MR (mean = 6.67 species/m2 ± 1.24 species/m2 SE) and Humbug CA (mean = 7.56 species/m2 ± 1.60 species/m2 SE; T-test, t ratio= -0.44, df = 9.41, P = 0.67).

  • The diversity of the sponge community did not differ between Cascade Head Marine Reserve (mean = 5.67 species/m2 ± 1.09 species/m2 SE) and Cavalier Comparison Area (mean = 10.17 species/m2 ± 2.15 species/m2 SE; T-test, t ratio= 1.87, df = 7.39, P = 0.10).

Figure 1. Average sponge biomass and diversity does not differ between Otter Rock Marine Reserve and Cape Foulweather Comparison Area (left) or between Redfish Rocks Marine Reserve and Humbug Mountain Comparison Area (right).

Figure 1. Average sponge biomass and diversity does not differ between Otter Rock Marine Reserve and Cape Foulweather Comparison Area (left) or between Redfish Rocks Marine Reserve and Humbug Mountain Comparison Area (right).

Q3. The community composition of sponge families did not differ between any of the site pairs sampled.

  • The community composition of sponge families was not significantly different between Otter Rock Marine Reserve and Cape Foulweather Comparison Area (ANOSIM, Global R = 0.06, P = 0.348, based on Bray-Curtis similarity on 4th root transformed sponge family volume). At the transect scale there is 35% similarity between sponge families between sites.

  • The community composition of sponge families was not significantly different between Redfish Rocks Marine Reserve and the Humbug Mountain Comparison Area (ANOSIM, Global R = -0.104, P = 0.786, based on Bray-Curtis similarity on 4th root transformed transect-scale sponge volume).

  • The community composition of sponge families was not significantly different between Cascade Head Marine Reserve and Cavalier Comparison Area (ANOSIM, Global R = 0.111, P = 0.165, based on Bray-Curtis similarity on 4th root transformed transect-scale sponge volume). However, transects shared only 7% similarity among all samples indicating that between transect differences in the sponge community, regardless of the sampling area, were high.

Figure 2. The community composition of sponge families does not differ significantly between any of the site pairs sampled. ANOSIM results are shown for Otter Rock Marine Reserve and Cape Foulweather Comparison Area (top left), Redfish Rocks Marine Reserve and Humbug Mountain Comparison Area (bottom left), and Cascade Head Marine Reserve and Cavalier Comparison Area (top right).

Figure 2. The community composition of sponge families does not differ significantly between any of the site pairs sampled. ANOSIM results are shown for Otter Rock Marine Reserve and Cape Foulweather Comparison Area (top left), Redfish Rocks Marine Reserve and Humbug Mountain Comparison Area (bottom left), and Cascade Head Marine Reserve and Cavalier Comparison Area (top right).

Q4. Different sponge genera are dominant in Cascade Head Marine Reserve versus Cavalier Comparison Area.

  • Four sponge families are the most dominant sampled (comprising >1% of the total sponge biomass collected) and constitute 99% of the Otter Rock Marine Reserve biomass and 95% of the Cape Foulweather Comparison Area biomass. Family volumes did not differ significantly between sites (ANOVA, P< 0.05).

  • Eight sponge families are the most dominant sampled (comprising >1% of the total sponge biomass collected) and constitute 98% of the biomass in Redfish Rocks Marine Reserve and 97% of the biomass in Humbug Mountain Comparison Area. One family, Tetillidae, accounted for greater than 1% of the total sampled community biomass and it only occurred in Humbug Mountain Comparison Area There were no significant differences in volume between sites for the eight most abundant families (ANOVA, P>0.05).

  • Eleven sponge genera are the most dominant sampled (comprising >1% of the total sponge biomass collected) and constitute 99.7% of the biomass in Cascade Head Marine Reserve and 96.7% of the biomass in Cavalier Comparison Area. While sponges from the genus Isodictya and Clathria were the clear dominants found in the marine reserve, Xestospongia and Mycale were the most common genus found in the comparison area.

Figure 3. Dominant sponge families are similar between Otter Rock Marine Reserve and Cape Foulweather Comparison Area (left) but are different between Redfish Rocks Marine Reserve and Humbug Mountain Comparison Area (right).

Figure 3. Dominant sponge families are similar between Otter Rock Marine Reserve and Cape Foulweather Comparison Area (left) but are different between Redfish Rocks Marine Reserve and Humbug Mountain Comparison Area (right).

Figure 4. There were no significant differences in the response ratios (biomass inside/biomass outside) in the volume of dominant sponge families between Otter Rock Marine Reserve and Cape Foulweather Comparison Area (A) or between Redfish Rocks Marine Reserve and Humbug Mountain Comparison Area (B).

Figure 4. There were no significant differences in the response ratios (biomass inside/biomass outside) in the volume of dominant sponge families between Otter Rock Marine Reserve and Cape Foulweather Comparison Area (A) or between Redfish Rocks Marine Reserve and Humbug Mountain Comparison Area (B).

3.3.2 Algae

Q1. There were significant differences in macroalgal biomass between Otter Rock and Cape Foulweather, but not between other site pairs.

  • The Otter Rock Marine Reserve site supports nearly 3x higher macroalgal biomass (mean = 1542 g/m2 ± 386 SE) than the Cape Foulweather Comparison Area (mean = 554 g/m2 ± 91 SE; T-test, t ratio= 2.92, df = 8.0, P = 0.02), largely due to the dominance of Bossiella orbigniana inside the reserve.

  • The Redfish Rocks Marine Reserve site supports nearly 4x the total biomass of macroalgae (mean = 469 g/m2 ± 200 SE) present in the Humbug Comparison Area, though the high variability at Redfish Rocks Marine Reserve makes this difference not significant (mean = 129 g/m2 ± 36 SE; T-test, t ratio= 1.62, df = 10.0, P = 0.14).

  • The Cascade Head Marine Reserve supports over 3x the total biomass of macroalgae (mean = 166.1 g/m2 ± 63.4 SE) present in the Cavalier Comparison Area (mean = 49.7 g/m2 ± 21.8 SE). This difference was not significant (T-test, t ratio= 2.09, df = 8.85, P = 0.067). Data analysis was based on log(biomass). It should be noted that contamination by bryozoans and hydroids did occur on some species, biasing the data.

Q2. Average algal diversity was not significantly different between any of the site pairs sampled.

  • The diversity of the macroalgal community did not differ between the Otter Rock Marine Reserve (mean = 29.0 species ± 5.1 SE) and Cape Foulweather Comparison Area (mean = 32.2 species ± 2.2 SE; T-test assuming unequal variance, t ratio= -0.57, df = 4.15, P = 0.60).

  • The diversity of the macroalgal community did not differ between the Redfish Rocks MR (mean = 39.1 species/m2 ± 5.5 SE) and Humbug Mountain Comparison Area (Fig. 10; mean = 34.4 species/m2 ± 6.9 SE; T-test, t ratio= 0.53, df = 10.0, P = 0.61).

  • The species richness of the macroalgal community did not differ between the Cascade Head Marine Reserve (mean = 16.0 species/m2 ± 2.0 SE) and the Cavalier Comparison Area (mean = 16.2 species/m2 ± 3.2 SE; T-test, t ratio= -0.07 3, df = 27.1, P = 0.95).

Figure 5. The top panels show biomass differences between marine reserves and comparison areas, and the bottom panels show diversity (species richness) differences. Otter Rock Marine Reserve had significantly more biomass than Cape Foulweather Comparison Area but there were no other significant differences in site pairs or in diversity.

Figure 5. The top panels show biomass differences between marine reserves and comparison areas, and the bottom panels show diversity (species richness) differences. Otter Rock Marine Reserve had significantly more biomass than Cape Foulweather Comparison Area but there were no other significant differences in site pairs or in diversity.

Q3. There were slight differences in the algal community composition between each site pair. Cascade Head Marine Reserve and Cavalier Comparison area were the least similar site pair, with differences driven by three large-bodied species.

  • Community composition of macroalgae does differ somewhat between quadrats sampled at Otter Rock Marine Reserve versus Cape Foulweather Comparison Areas (ANOSIM, Global R = 0.39, P = 0.001, based on Bray-Curtis similarity on 4th root transformed macroalgal biomass). At the transect-scale, 50% similarity is shared among both reserve and comparison area sites.

  • Marginal differences exist in the community composition of macroalgae between Redfish Rocks Reserve and Humbug Mountain Comparison Area (ANOSIM, Global R = 0.23, P = 0.001, based on Bray-Curtis similarity on 4th root transformed macroalgal biomass). At the transect-scale, 38% similarity is shared among both reserve and comparison area sites.

  • Slight differences exist in the community composition of macroalgae between Cascade Head Marine Reserve and the Cavalier Comparison Area (ANOSIM, Global R = 0.24, P = 0.004, based on Bray-Curtis similarity on 4th root transformed macroalgal biomass). At the transect-scale, 33% similarity is shared among both the reserve and comparison area sites (Figure 1). Three species contributed accounted for 20% of the dissimilarity in community composition at the transect scale between the reserve and the comparison area, Laminaria longipes, Pleurophycus gardneri, and Fryeela gardneri, in part because these are species with larger thalli who drive the biomass patterns.

Figure 6. ANOSIM results indicate that the macroalgal community composition does not significantly differ between Otter Rock Marine Reserve and Cape Foulweather Comparison Area (top) but does differ slightly between Redfish Rocks Marine Reserve and Humbug Mountain Comparison Area (bottom). There are also slight differences in community composition between Cascade Head Marine Reserve and Cavalier Comparison Area (not shown).

Figure 6. ANOSIM results indicate that the macroalgal community composition does not significantly differ between Otter Rock Marine Reserve and Cape Foulweather Comparison Area (top) but does differ slightly between Redfish Rocks Marine Reserve and Humbug Mountain Comparison Area (bottom). There are also slight differences in community composition between Cascade Head Marine Reserve and Cavalier Comparison Area (not shown).

Q4: Similar species were dominant between each site pair. Significant differences in species abundances were only observed between Otter Rock Marine Reserve and Cape Foulweather Comparison Area.

  • Otter Rock Marine Reserve and Cape Foulweather Comparison Area: 12 macroalgal species were the most dominant species sampled (comprising >1% of the total macroalgal biomass collected, summed across sampling site) and constituted 71% of the marine reserve biomass and 90% of the comparison area biomass. Biomass of six macroalgal species differed significantly between the reserve and comparison area (ANOVA, P< 0.05, quadrats means were pooled among transects). Caliarthron tuberculosum, Plocamium cartilagineum, and Desmarestia munda were more abundant inside the reserve, while Rhodymenia californica, Callophyllis flabellulata, and Fryeella gardneri were more abundant in the comparison area.

  • Redfish Rocks Marine Reserve and Humbug Mountain Comparison Area: 8 algal species are most dominant species sampled (comprising >1% of the total macroalgal biomass collected) and constitute 97% of the marine reserve biomass and 92% of the comparison area biomass. The biomass of these species did not differ significantly between sites.

  • Cascade Head Marine Reserve and Cavalier Comparison Area: Rhodymenia californica & pacifica, Cryptopleura farlowiana, Callophyllis flabellulata, and two kelp species, Pleurophycus gardneri, Laminaria longipes were the macroalgal species with greatest abundance among all sampled pooled. However, the biomass of these 5 species did not differ significantly between the reserve and the comparison area reflecting high variance in biomass at the transect scale, not the site scale.

Figure 7. There were significant differences in the response ratios (biomass inside/biomass outside) of the biomass of dominant macroalgal species between Otter Rock Marine Reserve and Cape Foulweather Comparison Area (A), but no significant differences between Redfish Rocks Marine Reserve and Humbug Mountain Comparison Area (B).

Figure 7. There were significant differences in the response ratios (biomass inside/biomass outside) of the biomass of dominant macroalgal species between Otter Rock Marine Reserve and Cape Foulweather Comparison Area (A), but no significant differences between Redfish Rocks Marine Reserve and Humbug Mountain Comparison Area (B).

Our survey likely undersampled the algal communities at Cascade Head Marine Reserve and Cavalier Comparison Area.

Species-accumulation curves predicting species richness through Chao2 and Jacknife extrapolation permutations show that the estimated species counts exceed the observed species counts, indicating that we likely under-sampled the total macroalgal diversity at Cascade Head Marine Reserve and Cavalier Comparison Area (Colwell and Coddington 1994). We did not perform this analysis for the other site pairs, so it is unknown whether we under-sampled those areas.

We documented the occurrence of three species of brown macroalgae that have never before been reported in Oregon waters.

  • Laminaria longipes was first discovered in Oregon by Gayle Hansen and Jim Golden in their Redfish Rocks survey of 2008, extending the southern limit of this species from the Juan de Fuca Strait in Washington to southern Oregon. This species is common in the low intertidal of Alaska where it forms dense beds of narrow (<2”) strap-shaped blades interconnected by rhizomes. The subtidal form of this species is morphologically different with blades that are broader and often deeply split. It has been thought to be extremely rare. However, the ODFW extractive survey at Redfish Rocks found this species (confirmed by DNA) to be by far the dominant mid-story kelp in the area. Although usually smaller, the upright blades could reach up to ~3 feet in height and ~2 feet in diameter. Up to 6 blades were seen forming from a single holdfast, and up to 70 blades were counted within a single ¼ m2 quadrat. Moreover, in one quadrat, the biomass reached 376 grams, the greatest of any seaweed in our study. Perhaps the most amazing thing about this species was that when disturbed (collected), it oozed huge quantities of polysaccharides (slime). We could always tell when it was present via this feature alone.

  • Desmarestia tabacoides/foliacea is an acid weed that sporadically occurred at Redfish Rocks Marine Reserve. Only young oval blades up to about 10 cm in height of this papery thin unbranched brown were discovered. The blades lysed quickly due to their acid content so even getting the young material vouchered was quite difficult. Although a new record for Oregon, the single bladed Desmarestia is known from the Juan de Fuca Strait and also from southern California, Mexico and Japan. To the north the species is referred to D. foliacea, a strap-shaped blade, reaching up to a foot in length and 2” in diameter. To the south, the species is called D. tabacoides, a broad, slightly undulate oval-shaped blade that can reach up to a meter in length. Interestingly, studies that would separate or combine the 2 species via their molecular sequences have not yet been carried out.

  • Syringoderma phinneyi is a very tiny 1-cell-thick fan shaped brown algal that we found epiphytic on filamentous red algae. Unlike its more common sister species, S. abyssicola, this species is only known from a few rare collections in BC and California. Named for Harry Phinney, a Professor of Algae at Oregon State University for more than 30 years, it is fitting that we discovered it in Oregon’s marine reserves.

3.3.3 Other Invertebrates

Q1. There were no significant differences in invertebrate phyla biomass or abundance in any of the site pairs sampled.

  • The total biomass of invertebrate phyla did not differ between Otter Rock Marine Reserve (mean = 668 g/m2 ± 294 SE) and Cape Foulweather Comparison Area (mean = 326 g/m2 ± 94 SE; Wilcoxon test, P = 0.39).

  • The total abundance of invertebrate phyla did not differ between Otter Rock Marine Reserve (mean = 578 indiv./m2 ± 113 SE) and Cape Foulweather Comparison Area (mean = 683 indiv./m2 ± 249 SE; Wilcoxon test, P = 1.0).

  • The total biomass of invertebrate phyla did not differ between Redfish Rocks Marine Reserve (mean = 98.4 g/m2 ± 31 SE) and Humbug Comparison Area (mean = 96.7 g/m2 ± 20 SE; Wilcoxon test, P = 0.87).

  • The total abundance of invertebrate phyla did not differ between Redfish Rocks Marine Reserve (mean = 412.7 indiv./m2 ± 121 SE) ad Humbug Mountain Comparison Area (mean = 399.1 indiv./m2 ± 54 SE; Wilcoxon test, P = 0.69).

Q2. Diversity was not measured for invertebrate phyla because observations were made for the same six phyla in all areas.

Figure 8. There were no significant differences in the biomass (top panels) or abundance (bottom panels) of invertebrates between Otter Rock Marine Reserve and Cape Foulweather Comparison Area (left) or between Redfish Rocks Marine Reserve and Humbug Mountain Comparison Area (right).

Figure 8. There were no significant differences in the biomass (top panels) or abundance (bottom panels) of invertebrates between Otter Rock Marine Reserve and Cape Foulweather Comparison Area (left) or between Redfish Rocks Marine Reserve and Humbug Mountain Comparison Area (right).

Q3. The community composition of invertebrate phyla did not differ between any of the site pairs sampled.

  • Community composition did not vary between quadrats sampled in the Otter Rock Marine Reserve versus Cape Foulweather Comparison Area (ANOSIM, Global R = 0.10, P = 0.086, based on Bray-Curtis similarity on 4th root transformed invertebrate biomass). At the transect-scale, 72% similarity is shared among both reserve and comparison area transects.

  • Community composition did not vary between quadrats sampled at Redfish Rocks Marine Reserve versus Humbug Mountain Comparison Area (ANOSIM, Global R = 0.06, P = 0.057, based on Bray-Curtis similarity on square root transformed invertebrate biomass, g/m2). At the transect-scale, 66% similarity is shared among both reserve and comparison area transects.

Figure 9. ANOSIM results indicate that the invertebrate phyla composition does not significantly differ between Otter Rock Marine Reserve and Cape Foulweather Comparison Area (top) or between Redfish Rocks Marine Reserve and Humbug Mountain Comparison Area (bottom).

Figure 9. ANOSIM results indicate that the invertebrate phyla composition does not significantly differ between Otter Rock Marine Reserve and Cape Foulweather Comparison Area (top) or between Redfish Rocks Marine Reserve and Humbug Mountain Comparison Area (bottom).

Q4: Significant differences in phyla biomass were only observed between Otter Rock Marine Reserve and Cape Foulweather Comparison Area, where Phylum Echinodermata was more abundant in the comparison area.

  • There was significantly more Echinodermata biomass in the Cape Foulweather Comparison Area compared to Otter Rock Marine Reserve. There were no significant differences in any of the other phyla (t-test on log10 transformed biomass; p > 0.05).

  • There are no significant differences in phyla biomass between Redfish Rocks Marine Reserve and Humbug Mountain Comparison Area.

Figure 10. Phlya biomasses were similar between Otter Rock Marine Reserve and Cape Foulweather Comparison Area (left) and between Redfish Rocks Marine Reserve and Humbug Mountain Comparison Area, with the exception of significantly higher biomass of Echinodermata in Cape Foulweather Comparison Area compared to Otter Rock Marine Reserve.

Figure 10. Phlya biomasses were similar between Otter Rock Marine Reserve and Cape Foulweather Comparison Area (left) and between Redfish Rocks Marine Reserve and Humbug Mountain Comparison Area, with the exception of significantly higher biomass of Echinodermata in Cape Foulweather Comparison Area compared to Otter Rock Marine Reserve.

3.4 Discussion and Considerations

The Benthic Extraction tool and collaboration with taxonomic experts generated important baseline information about the invertebrate, sponge, and macroalgal communities in Otter Rock Marine Reserve, Redfish Rocks Marine Reserve, Cascade Head Marine Reserve, and associated comparison areas. These data may be used to support future targeted studies on the invertebrate, macroalgal, or sponge communities in these areas. Benthic Extraction proved to be a costly endeavor, in terms of sampling equipment, ship-time, and personnel, that ultimately was not conducive to a regular monitoring schedule. While working closely with taxonomic experts proved to be a worthwhile collaboration (e.g., first documentation of 3 macroalgal species in Oregon), the identification of many groups to relevant taxonomic levels (e.g., species) is extremely time intensive and sometimes impossible without genetic confirmation.

4 Video Sled

4.1 Introduction

The video sled was part of a suite of visual survey tools, including the video lander, remotely operated vehicle (ROV) and SCUBA surveys, that were utilized in Otter Rock Marine Reserve and Redfish Rocks Marine Reserve in 2010 and 2011. These early efforts supported site characterization, systematic rapid assessment, and a detailed biological assessment of the two marine reserves.

The main objectives of the sled surveys were to:

  1. Determine the overall density of fish and invertebrates,
  2. Identify species-specific patterns of abundance for fish and invertebrates,
  3. Determine the community composition of fish and invertebrates, and
  4. Determine if a given reserve and its comparison areas differ in these metrics.

Substrate type and depth identified from seafloor mapping dictated which type of visual survey and equipment were used to collect these data. Video lander, remotely operated vehicle (ROV), and SCUBA divers were used to conduct visual surveys in areas of consolidated hard bottom. Video sled, video lander, and ROV were used to conduct visual surveys in unconsolidated sediment areas.

Video sleds are inexpensive and have been successfully used for surveying flat-bottomed areas for both fish and invertebrate assessments, with little damage to substrate and sessile organisms (Spencer et al. 2005). Both the Otter Rock and Redfish Rocks sites contain large areas of unconsolidated sediment. For these reasons we chose to conduct video sled surveys to support the detailed biological assessment of Otter Rock Marine Reserve and Redfish Rocks Marine Reserve.

The video lander and SCUBA surveys continued to be utilized as monitoring tools. However, due to high vessel costs needed to deploy the sled and the limited biological data generated by this tool, use of the video sled was ultimately discontinued.

4.2 Methods

4.2.1 Configuration

In its initial 2010 configuration, the sled frame was towed in contact with the bottom, with the lower frame rails acting as runners. A single video system was used with an inexpensive battery-powered Aqua-Vu unit which sent a black-and-white image via a 91-meter umbilical to the boat, where we recorded it on mini-DV videotape. Time, location, and depth data were collected every two minutes or as frequently as possible. The video was later analyzed and data on time, depth, location, habitat type, and organisms observed were entered into an Excel spreadsheet.

In 2011, a new sled frame was constructed, with the goal of having the sled frame float at a controllable distance off-bottom to adapt to changing relief. “Dropper chains” were used to pull the sled down until approximately half the chain was supported on-bottom, allowing the sled frame to settle into a neutrally buoyant state a short, controllable distance off-bottom. The height of the frame above bottom could be adjusted by varying the length of line connecting the frame to the top link of dropper chains. In practice, we were able to successfully tow the new sled system over long tracts of hard bottom, including instances of high-relief benthic structure. The video systems were revised at the start of the 2011 season. We incorporated the autonomous underwater video system or “tube” system previously used on the lander. This system used an aluminum pressure tube to house a Sony mini-DV camcorder and batteries, and cables ran out to a Deep Sea Power & Light (DSPL) 2060 low-light color camera, DSPL Rite-lite light with 5-watt LED flood, and DSPL parallel red lasers with 10-cm spacing to estimate scale. The autonomous system allowed for high quality standard-definition color video to be collected for later analysis, and used a Horita PG-2100 time-code generator synced to an onboard GPS unit collecting location data, so that the video could be accurately geo-referenced (Tissot 2008). The system also included a wired underwater Sea-Viewer camera with 183 m (600 ft) umbilical cable connected to a 12-volt battery and Sony GV-HD700 video recording deck on the boat. This camera delivered a live video image to the boat that showed the sled frame attitude and status as well as potential obstacles in the tow path of the sled. We used the Garmin 546 chartplotter, ToughBook computer, and Fugawi software to collect track data for the sled. We defined a “track” as a record, taken once per second, of Greenwich mean time and the latitude/longitude and UTM location along the length of a transect.

4.2.2 Study Design

Survey months and number of transects conducted in 2010 and 2011 at the Otter Rock site and the Redfish Rocks site are provided in Table 1. In 2010, our video sled survey transects totaled 3,551 meters at the Otter Rock site and 8,049 meters at the Redfish Rocks site (Figure 11, left). In 2011 our video sled survey transects totaled 5,057 meters at the Otter Rock site and 5,553 meters at the Redfish Rocks site (Figure 11, right).

For detailed biological assessments, the video sled was deployed in a stratified random design within the Otter Rock and Redfish Rocks sites. Random points were assigned within unconsolidated substrates binned by depth (0-7, 7-14, 14-21, 21+ meters). The random points within each depth stratum were used as starting points for sled tows. Once the boat was on-station, we would generally drift down-current while making final preparations and then turn into the current to begin the tow. Optimal towing speed was 0.5 – 0.8 meters per second (1-1.5 knots). We aimed for a 20-30 minute tow time, yielding a tow length of 700-1,000 meters. The topside video monitor was continually observed for sled safety. Once the appropriate on-bottom time had elapsed, or if inappropriate habitat was encountered, the sled was retrieved and hauled aboard using a crab block.

Figure 11. Video sled transects conducted in 2010-2011 for the Otter Rock site (left) and Redfish Rocks site (right).

Figure 11. Video sled transects conducted in 2010-2011 for the Otter Rock site (left) and Redfish Rocks site (right).

Table 1. Video sled surveys conducted at the Otter Rock and Redfish Rocks sites.

Table 1. Video sled surveys conducted at the Otter Rock and Redfish Rocks sites.

4.2.3 Data Extraction

We developed a standard method to examine the video from each sled transect. Video was captured using Adobe Premiere software and examined for sled status, substrate type, and biotic community. We overlaid a mask on the video screen, superimposing two horizontal lines: one corresponding to 80% of the vertical distance from the bottom of the screen and another at 50%, roughly the level where the lasers struck the substrate. Primary substrate was defined as comprising >50% of the video screen from the 50% line to the bottom of the screen; secondary substrate was defined as comprising 20-50% of the same area. Fish in the water column were counted when any part of the fish passed below the 80% line, constraining the survey to a fixed width and accounting for the practical limits of underwater visibility (Amend et al. 2001; Donnellan et al. 2008). Invertebrates, macroalgae, and other bottom-dwellers were counted as they passed below the 50% line.

In 2010, the sled was towed touching the benthos, therefore the transect area (m2) was calculated as a fixed width based on the distance between the rails of the sled, visible within the field of view, when sled was on bottom (0.88 m) x transect length (provided by the 2010 Access database based on handheld GPS points taken approximately every two minutes). Organisms were scored in 10 second segments over the course of the transect. These segments were summed to generate a total abundance per organism per transect. Organisms were assumed to be observed “on bottom” as no mention of ascent/descent was made in the database. Abundances were then divided by the transect area to generate relative organism density (indiv./m2) per transect.

In 2011, the sled was towed slightly above the benthos. Transect length was calculated using the Fugawi (navigational software) tracks which recorded a coordinate every 2-3 seconds. These tracks represent the maximum transect length (including time to descend and ascend the sled from the ship). In order to restrict the transect length to “on bottom” time only, the tracks were segmented to the first occurrence of habitat scoring when we assume bottom contact was made. Likewise, “gaps” exist where the sled loses view of the benthos rendering that section ‘un-scorable’ for habitat and organisms. Both the gaps and pre-transect and post-transect track lines (during ascent and descent) were excluded from calculating the adjusted transect length (m). Transect width (m) was estimated from repeated measures of the width of field of view using the two fixed parallel 10-cm width lasers as a reference point which appear near the middle of the reviewing screen. The estimated transect area (m2) was then calculated from the adjusted transect length x the estimated width. Organisms were scored continuously over the course of the transect and summed to generate a total abundance per organism per transect. Organisms were queried to exclude those observed on descent or ascent. Observations of “Unknown Species” were observed but excluded from the analysis (resulting in 144 individual organisms excluded from the analysis). Organism abundances were divided by the transect area to generate relative organism density (indiv./m2) per transect.

Each transect serves as a single replicate, irrespective of total area surveyed. Mean depth (m) per transect was calculated from the depth data collected from the vessel’s sounder periodically (2010) or every two minutes (2011). If no depth data was recorded from field collection (e.g., select transects in 2010), depth data was extracted from a bathymetric raster layer in ArcGIS based on the spatial position of the transect.

The quality control procedure for sled video review was similar to that conducted for the video lander. Twenty percent of the 2011 video sled transects were randomly selected and reanalyzed by a second observer. Fish and invertebrates were summed per transect and across all transects within a site and compared between reviewers to verify consistency in species identifications. All database entries were error-checked against the original paper copies for entry error and corrected.

4.2.4 Data Analysis

Data analysis was conducted separately for Redfish Rocks Marine Reserve and associated comparison areas (henceforth Redfish Rocks area) and Otter Rock Marine Reserve and associated comparison areas (henceforth Otter Rock area). We grouped organisms into three main groups: fish, mobile invertebrates, and sessile invertebrates. For each organism group, we assessed patterns in aggregate density, species-specific densities, and community composition.

ANCOVA analysis (continuous covariate = mean transect depth, factor = sampling area) was used to compare aggregate fish, mobile invertebrate, and sessile invertebrate densities (indiv./m2) among the reserve, MPA, and comparison areas while exploring the influence of depth (m). Aggregate fish, mobile invertebrate, and sessile invertebrate densities in the Redfish Rocks area were log10 transformed to achieve normality. In the Otter Rock area, aggregate mobile invertebrate density was also log10 transformed, but aggregate fish density was approximately normally distributed and did not require transformation. Finally, sea anemones were the only sessile invertebrate observed in the Otter Rock area. Mean densities of sea anemones were compared using a non-parametric Wilcoxon test due to the lack of normality and homoscedasticity in the density data.

Species-specific densities were compared using a non-parametric Kruskal-Wallis (Redfish Rocks area) or Wilxocon test (Otter Rock area) due to the lack of normality and homoscedasticity in the data.

ANOSIM (factor = sampling area) was used to explore differences in community composition. However, species-specific identification was limited during the 2010-11 sled tows. Without species-specific resolution in the density data, exploring community composition among the marine reserve and comparison sites is not informative. Therefore, in the following section we present ANOSIM results for the only organism group with sufficient species-specific data: mobile invertebrates in the Redfish Rocks area. Density data were first square root transformed, then Bray-Curtis similarity was calculated for each transect.

4.3 Results

4.3.1 Redfish Rocks Marine Reserve

4.3.1.1 Fish

4.3.1.1.1 Aggregate density

Total fish density did not vary with mean transect depth (ANCOVA; P = 0.93), nor was there a significant interaction between depth and sampling area (P = 0.47). There was no difference in mean fish density between Redfish Rocks Marine Reserve (mean = 0.083 ± 0.041 SE), the MPA (mean = 0.015 ± 0.007 SE), Humbug Comparison Area (mean = 0.057 ± 0.009 SE), and McKenzie Reef Comparison Area (mean = 0.012 ± 0.011 SE; ANCOVA, P = 0.86).

4.3.1.1.2 Species-specific densities

No significant differences in species-specific fish densities were detected between the reserve and comparison area. The fish observations in both the MPA and McKenzie Comparison Area consisted only of flatfish likely due to the lack of hard substrates encountered by the sled in these two areas.

4.3.1.1.3 Community composition

The majority of the fish observations were categorized as unidentified fish, flatfish, or rockfish. Once unidentified fishes were excluded from the analysis, only eight of the 24 sled transects contained species-specific fish observations. Hence, ANOSIM analyses were not conducted.

4.3.1.2 Invertebrates

4.3.1.2.1 Aggregate density

Total mobile invertebrate density did not vary with mean transect depth (ANCOVA; P = 0.92), nor was there a significant interaction between depth and sampling area (P = 0.76). There was no significant difference in mean mobile invertebrate density between Redfish Rocks Marine Reserve (mean = 0.430 ± 0.15 SE), MPA (mean = 0.037 ± 0.017 SE), the Humbug Comparison Area (mean = 0.098 ± 0.038 SE), and the McKenzie Reef Comparison Area (mean = 0.091 ± 0.057 SE; ANCOVA, P = 0.27). There was no significant difference in mean sessile invertebrate density between Redfish Rocks Marine Reserve (mean = 0.114 ± 0.04 SE), the MPA (mean = 0.004 ± 0.001 SE), the Humbug Comparison Area (mean = 0.050 ± 0.028 SE), and the McKenzie Reef Comparison Area (mean = 0.023) P = 0.21).

4.3.1.2.2 Species-specific densities

No significant differences in species-specific mobile invertebrate densities were detected between the Redfish Rocks Marine Reserve, MPA, and comparison areas. However, sea pens (a sessile invertebrate) differed significantly in abundance between the Redfish Rocks MPA and Humbug Comparison Area (P = 0.024). No other significant differences in sessile invertebrate densities were detected between the reserve and comparison area.

4.3.1.2.3 Community composition

ANOSIM did not reveal any significant grouping of the observed mobile invertebrate communities among the four sampling areas (P = 0.21; Global R: 0.065). Species-specific identification of sessile invertebrates was too low to conduct ANOSIM analyses for this group.

4.3.2 Otter Rock Marine Reserve

4.3.2.1 Fish

4.3.2.1.1 Aggregate density

Total fish density did not vary with mean transect depth (P = 0.28), nor was there a significant interaction between depth and sampling area (P = 0.80). There was no difference in total fish density between Otter Rock Marine Reserve (mean = 0.007 ± 0.002 SE) and the Moolack comparison area (mean = 0.014 ± 0.010 SE; ANCOVA, P = 0.67).

4.3.2.1.2 Species-specific densities

Mean densities of individual fish species or species groups were compared using non-parametric Wilcoxon test due to the lack of normality and homoscedasticity in the density data. No significant differences in fish densities were detected between the reserve and comparison area.

4.3.2.1.3 Community composition

Species-specific fish identification was too low to conduct ANOSIM analyses for this group. The majority of the fish observations were categorized as unidentified fish, flatfish, or rockfish. In addition, fish abundances were very low resulting in several tows observing a single fish over the duration of the transect.

4.3.2.2 Invertebrates

4.3.2.2.1 Aggregate density
Figure 12. Regression relationship between transect depth and mobile invertebrate density (y-axis on log10 scale). Red circles are transects completed in the Otter Rock Marine Reserve (MR); blue circles are transects completed in the Moolack comparison area (CA). A single linear regression line is shown, as there was no difference between the two sampling areas (ANCOVA analysis).

Figure 12. Regression relationship between transect depth and mobile invertebrate density (y-axis on log10 scale). Red circles are transects completed in the Otter Rock Marine Reserve (MR); blue circles are transects completed in the Moolack comparison area (CA). A single linear regression line is shown, as there was no difference between the two sampling areas (ANCOVA analysis).

Total mobile invertebrate density did vary with mean transect depth (ANCOVA; P = 0.002), such that deeper depths correlated with higher relative densities of mobile invertebrates irrespective of sampling site (Figure 12). There was no difference in mean mobile invertebrate density between Otter Rock Marine Reserve (mean = 0.023 ± 0.014 SE) and the Moolack Comparison Area (mean = 0.145 ± 0.043 SE; ANCOVA, P = 0.12).

Sea anemones were the only sessile invertebrate observed in Otter Rock sled transects. No significant differences were detected between the reserve (mean = 0.005 ± 0.004) and comparison area (mean = 0.001 ± 0.001).

4.3.2.2.2 Species-specific densities

Sand dollars (Dendraster spp.) were significantly more abundant within Moolack CA while sea stars and brittle stars were significantly more abundance in the reserve. No other significant differences in mobile invertebrate densities were detected between the reserve and comparison area.

4.3.2.2.3 Community composition

Species-specific invertebrate identification was too low to conduct ANOSIM analyses for this group. Additionally, low densities (i.e. often a single organism type was observed during the entire transect), make a community composition analysis unreasonable.

4.4 Discussion and Considerations

Underwater visual surveys revealed very low densities of nearshore fishes with the sled tool. The low densities combined with poor species-specific identification severely hindered the assessment of fish community composition between the marine reserve and comparison areas. However, those fishes observed were more common on rugose, consolidated substrates (i.e. bedrock outcrops and boulders). Using the video sled, the abundance and community composition of mobile invertebrates were not found to differ between the reserve and comparison areas. Similarly, few differences were found in the abundance of sessile invertebrates, with the exception of sea pens being more abundant in the Redfish Rocks MPA compared to the Humbug Comparison Area. Comparison of community composition for sessile invertebrates was not informative with the limited data.

The video sled did not generate informative biological data relative to other visual survey tools. For example, while the sled surveys did not identify significant differences in invertebrate abundance and community composition (except for sea pens), SCUBA surveys revealed significant differences in several invertebrate taxa including red urchins and white plumed anemone between Redfish Rocks Marine Reserve and associated comparison areas (ODFW, 2014). In addition, at Otter Rock Marine Reserve and associated comparison areas, SCUBA surveys were more easily able to explicitly target rugose reef habitats that had higher abundance and diversity of fish and invertebrates.

Our work with the video sled represented an important exploration of a suite of visual survey techniques during the baseline assessment of Otter Rock Marine Reserve, Redfish Rocks Marine Reserve, and associated comparison areas. From this experience, we were able to fine tune our approach and continue with only the tools that yielded reliable data with high taxonomic resolution. Despite a concerted effort to develop the video sled into a monitoring tool (e.g., database creation, rigorous quality control methods), including using lessons learned during the first year to adapt sled configuration in the next sampling year (e.g., dropper chain, updated video systems), the video sled tool was discontinued due to the high vessel costs needed to deploy the sled and limited biological data generated.

5 Otolith Sampling

5.1 Introduction

All teleost fishes have otoliths, which are calcium carbonate inner ear stones that aid in fish orientation. Otoliths have long been an important fisheries tool used to examine the age distributions of fish populations because material is added to the otolith daily and annual bands are formed. Measurement of the age and length distributions of economically important fish species can be a way to monitor the response of a population to fishing pressure and marine reserve protection. For example, a possible result of fishing pressure is the truncation of the age distributions of populations, as larger, older fishes are often harvested first (e.g., Jorgensen et al., 2007). Increased mortality from fishing is also expected to favor faster life histories, potentially reducing age at reproductive maturity (Heino et al., 2015). Protection from fishing pressure can also increase the number of large, old, fecund female fish (Hixon et al., 2014). For these reasons, we might expect shifts in age distributions, and changes in age-length relationships, as a result of protection over time. We might also expect these shifts to differ by sex.

The time it takes to observe a shift in the age distribution of a population after changes in harvest regulations depends on the life history of the species (e.g., recruitment variability, age to maturity); however, a study in a comparable temperate marine reserve system with many of the same species or congeners observed significantly larger fishes in a marine reserve relative to a comparison area after 20 years of protection (Starr et al., 2020). Our otolith sampling occurred before harvest restrictions began. Therefore, the data collected serve as baseline information about age distributions and age-length relationships at several sites. Because Black Rockfish are often the most abundant species caught during Hook-and-Line (HnL) surveys, they were selected as a focal species for otolith sampling to maximize sample size.

The main objectives of otolith sampling were to:

  1. Determine the age distribution of Black Rockfish at each site,
  2. Determine the age-length correlation of Black Rockfish at each site, and
  3. Identify any differences between sites (marine reserve versus comparison area) and by sex.

5.2 Methods

Black Rockfish otolith samples were collected during HnL surveys at Redfish Rocks Marine Reserve and Humbug Comparison Area in 2011 and at Cascade Head Marine Reserve, Schooner Creek Comparison Area, Cape Perpetua Marine Reserve and Postage Stamp Comparison Area in 2013.

All otolith pairs were aged using two methods: surface and burn/bake break aging methods. All otoliths were soaked in a 50% ethanol solution at least overnight to improve pattern clarity. After soaking, one of the otoliths was examined whole (surface) to determine an approximate age and the other was broken and burned/baked for comparison. At times the surface age of an otolith was much easier to interpret than the burned or baked half (usually in young fish <6 years).

All aging work was conducted by a single reader, Lisa Kautzi, a marine age reading specialist with the ODFW Marine Resources Program. Ages from the two aging methods were averaged for this analysis.

5.3 Results

5.3.1 Redfish Rocks Marine Reserve

We analyzed otoliths from 76 female Black Rockfish. The age range was 4-14.5 yrs and the length range was 26-49 cm. We analyzed 75 male Black Rockfish. The age range was 2.5–19.5 yrs and the length range was 28-48 cm. Males are larger from 0-6 years; females diverge from males by year 6 and are a larger size for each age from 6-14 years (Figure 13A). Oldest male samples range in age from 13-19.5 yrs and in length from 33-43 cm which is up to 10 cm below the trend line.

Figure 13. Age-length correlation for male (blue) and female (orange) Black Rockfish (A) and their length distribution (B). Samples were collected from Redfish Rocks Marine Reserve.

Figure 13. Age-length correlation for male (blue) and female (orange) Black Rockfish (A) and their length distribution (B). Samples were collected from Redfish Rocks Marine Reserve.

5.3.2 Humbug Comparison Area

We analyzed otoliths from 99 females. The age range was 3-23.5 yrs and the length range was 26-52 cm. The oldest female is 23.5 years and 38.5cm, which is approximately 12cm below the length at age trend line. We analyzed otoliths from 77 males. The age range was 3-17 yrs and the length range was 26-46 cm. The oldest male samples are from 1-5 cm below the trend line and range from 14-17 yrs. Male are larger than females between 0-5 yrs. Females diverge from males at 5 yrs and are larger at any age above 6.5yrs (Figure 14A).

Figure 14. Age-length correlation for male (blue) and female (orange) Black Rockfish (A) and their length distribution (B). Samples were collected from Humbug Comparison Area.

Figure 14. Age-length correlation for male (blue) and female (orange) Black Rockfish (A) and their length distribution (B). Samples were collected from Humbug Comparison Area.

5.3.3 Cascade Head Marine Reserve

We analyzed otoliths from 65 females. The age range was 5-11.5 yrs and the length range was 34-51 cm. Four of the five oldest females (10-11.5 yrs) are smaller than the trend line would predict by 1-4 cm. We analyzed otoliths from 75 males. The age range was 3.75-13.5 yrs and the length range was 28-47 cm. Two of the oldest males are 1-2 cm above the trend line, while the oldest is 1 cm below the trend line. The divergence point between females and males is approximately 5 yrs, with females generally larger than males (Figure 15A). Female samples are tightly grouped at 5 yrs. and have we did not have any younger samples to anchor the trend line at less than 5 yrs.

Figure 15. Age-length correlation for male (blue) and female (orange) Black Rockfish (A) and their length distribution (B). Samples were collected from Cascade Head Marine Reserve.

Figure 15. Age-length correlation for male (blue) and female (orange) Black Rockfish (A) and their length distribution (B). Samples were collected from Cascade Head Marine Reserve.

5.3.4 Schooner Creek Comparison Area

We analyzed otoliths from 27 females. The age range was 4.5-16.5 yrs and the length range was 40-55 cm. The oldest female is 16.5 yrs and is 1 cm above the trend line at 55 cm. Females tend to be slightly larger than their male counterparts at the same age. We analyzed otoliths from 30 males. The age range was 5-14 yrs. The length range was 29-46 cm. The oldest males at 14 yrs. are equal or 1 cm below the trend line. The divergence point between females and males is not clearly distinguish because there is only a single sample below age 5 (Figure 16A).

Figure 16. Age-length correlation for male (blue) and female (orange) Black Rockfish (A) and their length distribution (B). Samples were collected from Schooner Creek Comparison Area.

Figure 16. Age-length correlation for male (blue) and female (orange) Black Rockfish (A) and their length distribution (B). Samples were collected from Schooner Creek Comparison Area.

5.3.5 Cape Perpetua Marine Reserve

We analyzed otoliths from 87 females. The age range was 2.75-10.5 yrs and the length range was 25-51 cm. The oldest females are 10.5 yr and are 49 and 51 cm. A 5 yr female is also 51 cm long. We analyzed otoliths from 65 males. The age range was 3.25-16.5 yrs and the length range was 28-49. The oldest male is 3 cm below the trend line at 44 cm. The divergence point between females and males is less than 4 yrs, the earliest divergence seen in the entire data set (Figure 17A).

Figure 17. Age-length correlation for male (blue) and female (orange) Black Rockfish (A) and their length distribution (B). Samples were collected from Cape Perpetua Marine Reserve.

Figure 17. Age-length correlation for male (blue) and female (orange) Black Rockfish (A) and their length distribution (B). Samples were collected from Cape Perpetua Marine Reserve.

5.3.6 Postage Stamp Comparison Area

We analyzed otoliths from 79 females. The age range was 3-11 yrs and the length range was 26-50 cm. The oldest female anchors the trend line at 11 yrs and 50 cm. We analyzed otoliths from 90 males. The age range was 2-12.5 yrs and the length range was 21-51 cm. The five oldest males are between 11-12.5 yrs and fit nearly equi-distantly around the trend line. The divergence point between females and males is 5 yrs (Figure 18A).

Figure 18A. Age-length correlation for male (blue) and female (orange) Black Rockfish (A) and their length distribution (B). Samples were collected from Postage Stamp Comparison Area.

Figure 18A. Age-length correlation for male (blue) and female (orange) Black Rockfish (A) and their length distribution (B). Samples were collected from Postage Stamp Comparison Area.

5.4 Discussion and Considerations

Our otolith sampling represented an effort to develop baseline information about the age distributions and age-length relationships for Black Rockfish at six areas including Redfish Rocks Marine Reserve, Humbug Comparison area, Cascade Head Marine Reserve, Schooner Creek Comparison Area, Cape Perpetua Marine Reserve and Postage Stamp Comparison Area. We analyzed ~ 850 otoliths in total. We found that the oldest individuals were caught in the Redfish Rocks Marine Reserve and Humbug Comparison Area and the youngest individuals were caught in the Cape Perpetua Marine Reserve and Postage Stamp Comparison Area. Additionally, we found that male and female age-length relationships diverge between 4-6 years, depending on location. In each location, the slope of the age-length relationship was slightly greater for females than for males after the divergence point, indicating that females reach larger sizes at a given age than males after 4-6 years. Unfortunately, limited personnel time prevented us from statistically analyzing differences in age distributions and age-length relationships between sites and by sex.

Despite the information generated in this study, otolith sampling was ultimately discontinued. A power analysis indicated that, given the small size of the reserves and other ecological factors (e.g., degree of population mixing between sites), the sample sizes that would be necessary to capture site-based differences in age distribution or age-length relationships over time were prohibitively large. Thus, otolith sampling was determined not to be an appropriate tool for the Marine Reserves Program’s ecological monitoring goals. However, the methodology and lessons learned in this study could enable future otolith studies for future, more targeted research questions.

6 References

Amend M, Fox D, and Romsos C (2001) Nearshore rocky reef assessment, ROV survey. Oregon Department of Fish and Wildlife, 1-30

Dayton PK (1971) Competition, disturbance, and community organization: The provision and subsequent utilization of space in a rocky intertidal community. Ecol Monogr 41(4): 351-389

Donnellan M, Merems A, and Miller W (2008) Developing fish monitoring methods on southern Oregon reefs. Final Report to the Department of Land Conservation and Development in Fulfillment of 2006-07 Coastal Zone Management Section 309 Grant, Cooperative Agreement PS06005: 1-43

Heino M, Diaz Pauli B, and Dieckmann U (2015). Fisheries-induced evolution. Annual review of ecology, evolution, and systematics 4(46): 461-80.

Hixon MA, Johnson DW, and Sogard SM (2014). BOFFFFs: on the importance of conserving old-growth age structure in fishery populations. ICES Journal of Marine Science 71, 2171–2185.

Jorgensen C, Enberg K, Dunlop ES, Arlinghaus R, Boukal DS, Brander K, Ernande B, Gardmark A, Johnston F, Matsumura S, Pardoe H (2007). Ecology-Managing evolving fish stocks. Science 318(5854): 1247-8

Medina M, Andrade S, Faugeron S, Lagos N, Mella D, Correa J (2005) Biodiversity of rocky intertidal benthic communities associated with copper mine tailing discharges in northern chile. Mar Pollut Bull 50(4): 396-409

ODFW (2014). Oregon Marine Reserve Ecological Monitoring Report 2010-2011. Oregon Department of Fish and Wildlife. Marine Resources Program. Newport Oregon. 1-131.

Pringle J (1984) Efficiency estimates for various quadrat sizes used in benthic sampling. Can J Fish Aquat Sci 41(10): 1485-1489

Spencer ML, Stoner AW, Ryer CH, Munk JE (2005) A towed camera sled for estimating abundance of juvenile flatfishes and habitat characteristics: Comparison with beam trawls and divers. Estuar Coast Shelf Sci 64(2): 497-503

Starr RM, Wendt DE, Barnes CL, Marks CI, Malone D, Waltz G, et al. (2015) Variation in Responses of Fishes across Multiple Reserves within a Network of Marine Protected Areas in Temperate Waters. PLoS ONE 10(3): e0118502.

Tissot BN. 2008. Video analysis, experimental design, and database management of submersible-based habitat studies. In: Reynolds JR and Greene HG (eds) Marine habitat mapping technology for Alaska. Alaska Sea Grant College Program, University of Alaska Fairbanks, pp 157