1 Introduction: Cascade Head ROV Invertebrate Survey Report

The remotely operated vehicle (ROV) is our most complex monitoring tool. The ODFW Marine Reserves program partners with the ODFW Marine Habitat Project, a partner research group within ODFW to conduct this type of monitoring in the marine reserves. The ROV is driven by an operator from a boat, controlled via an umbilical cable. The ROV can swim up, down, and around obstacles and follow along a transect line, like a SCUBA diver. It collects high-definition video that is later used to analyze fish, invertebrates and benthic habitat structure within the marine reserve and its associated comparison areas. The ROV is perfect for surveying rocky habitats all the way out to the deepest parts of the reserves.

ROV surveys were initiated prior to reserve closure at the Cascade Head Marine Reserve in 2013, one year before harvest restrictions began. Sampling is conducted in the marine reserve and its associated comparison areas, Cavalier and Schooner Creek (see methods Appendix for additional information about comparison area selection). We sampled at these sites over several years, with varied levels of success in achieving usable data - data that met requirements for view, visibility, and benthic habitat type (rocky substrates). These efforts results in three years of usable data for our analysis and inclusion in the synthesis report.

Data from ROV monitoring efforts can be used to explore questions about invertebrate relative abundance from a non-extractive, fisheries-independent tool used to survey other deep reefs off the Oregon and the US West Coast. We can use metrics for diversity and community composition derived from these data to compare across monitoring tools, to understand tool bias, or to validate trends in relative abundance observed across tools. Data on relative abundance also enables us to explore how invertebrate communities change over time; and whether these changes are similar both inside the reserve and outside in comparison areas. For all data our main focus is exploring trends by site and year.

1.1 Survey Maps

1.1.1 Cascade Head Marine Reserve

Fig. 1: Map of ROV transects at the Cascade Head Marine Reserve

Fig. 1: Map of ROV transects at the Cascade Head Marine Reserve

1.1.2 Schooner Creek Comparison Area

Fig. 1: Map of ROV transects at the Schooner Creek Comparison Area

Fig. 1: Map of ROV transects at the Schooner Creek Comparison Area

1.1.3 Cavalier Comparison Area

Fig. 1: Map of ROV transects at the Cavalier Comparison Area

Fig. 1: Map of ROV transects at the Cavalier Comparison Area

1.2 Research Questions

Diversity

  • Does species diversity vary by site or year?

Community Composition

  • Does community composition vary by site or year?
    • If yes, what species drive this variation?

Aggregate Abundance

  • Does aggregate density vary by site or year?

Focal Species Abundance

  • Does focal species density vary by site or year?
  • Does focal species size vary by site or year?

2 Takeaways

Here we present a summary of our ROV invertebrate monitoring results and our conclusions. Our conclusions are written with an evaluation of our sampling design, knowledge from prior marine reserves monitoring reports, and future directions of marine reserves monitoring in mind.

2.1 ROV Invertebrate Results Summary

Invertebrate species diversity was similar between Cascade Head Marine Reserve and its associated comparison areas.

Species invertebrate diversity is similar between the Cascade Head Marine Reserve and Schooner Creek and Cavalier Comparison Areas as evidenced by the results of multiple analyses in the diversity section of this report. They have similar numbers of estimated and observed species, as well as similar numbers of unique and common species. While there were some differences in the Hill Diversity indices among sites, indicating slight differences in the concepts of species evenness among sites.

Invertebrate community composition differed minimally by site but varied more substantially by year.

The most substantial among-year differences in invertebrate community composition were seen between 2012 and 2017, with 2013 intermediate, consistent with a progression over time in the multivariate composition of the invertebrate community. The top four species with the strongest correlations to community composition, Burrowing Sea Cucumbers, California Sea Cucumbers, Fish Eating Anemones, and Basket Stars, all exhibited increasing densities over the sampling period at least one site. Differences among sites in overall community composition were minimal. Variability in invertebrate community composition among transects was highest at the Cavalier Comparison Area and the lower at the Cascade Head Marine Reserve and the Schooner Creek Comparison Area.

Notable trends at the aggregate level for invertebrates include reductions in densities of sea stars and sponges, and an increase in sea cucumbers.

The most notable trends in aggregate density by taxonomic group were an overall decline in sea stars (associated with the Sea Star Wasting Syndrome that affected sea stars across the west coast) and sponges. There was also an increase in sea cucumbers (both Burrowing Cucumbers and California Sea Cucumbers) at two of the three sites. The causes for the decline in sponges and increases in sea cucumbers are unclear.

Where species density differed between the marine reserve and its comparison areas, the marine reserve had higher densities in four of six cases.

For four species (Red Sea Urchins, Rock Scallops, California Sea Cucumbers, and Basket Stars), densities were higher across years at the Cascade Head Marine Reserve than at least one of the two comparison areas. Densities were only lower at the marine reserve for the Sunflower Star (which disappeared from all sites by 2017), and for the Burrowing Sea Cucumber (for which that difference did not hold across all years). Two other species (White Plumose Anemones and Fish Eating Anemones) had non-significant trends toward lower density at the marine reserve than at least one comparison site. Only the Burrowing Cucumber exhibited a Site * Year interaction, indicating differing changes in density over time at different sites.

By 2017, we detected density increases for at least one site for six species, and only two species decreased significantly in density.

Six analyzed species (Red Sea Urchins, Rock Scallops, California Sea Cucumbers, Burrowing Sea Cucumbers, Fish Eating Anemones, and Basket Stars) exhibited increases in density for at least one site by 2017. Short Red Gorgonians and Sunflower Sea Stars were the only species to exhibit significant decreases in density by 2017. The increase in Red Sea Urchins at two sites, and the disappearance of Sunflower Sea Stars from all sites were dramatic changes reflecting yearly trends that were also observed at the Redfish Rocks Marine Reserve and associated comparison areas.

Invertebrate density relationships with depth varied by site but were predominantly increasing.

Species densities were influenced by relationships with depth to varying degrees and were mostly site specific. However, the predominant responses were either density increases with depth or unimodal responses in which density increased to a maximum value and declined again at the deepest depths. Of the species analyzed, there were no species that significantly declined in density with increasing depth across the range of depths sampled.

Differences among years in sampling depth and season may have affected the observed density trends and should be considered in future sampling efforts.

Differences in sampling depth among years may have influenced the observed density patterns. Further analysis of the existing data relative to depth may help identify individual species and depth ranges for which sampling stratification by depth should be pursued. Even though most invertebrates sampled are sessile, further assessment of potential seasonal effect is warranted.

Few invertebrates responded significantly to the percentage of boulder habitat.

Red Sea Urchins and Basket Stars had an inverse linear relationship with the proportion of boulder habitat across sites. White Plumose Anemones, Burrowing Sea Cucumbers, and Fish Eating Anemones displayed an assortment of non-linear relationships with the percentage of boulder substrate at individual sites.

2.2 Conclusions

This is the first ecological monitoring report to summarize ROV invertebrate data from the Cascade Head Marine Reserve and its associated comparison areas.

Despite completion of the first ROV surveys in 2012, this report provides the first summary of ROV invertebrate monitoring data at the Cascade Head Marine Reserve. This report documents the general similarity of the marine reserve to its two comparison areas – Cavalier and Schooner Creek. From a diversity and community composition perspective the differences among sites are minimal relative to changes over time. From an abundance perspective, some species exhibited density differences between the Cascade Head Marine Reserve and one or both comparison area, with the marine reserve generally having the higher densities.

Changes in the selection of comparison area sampling locations complicated the detection and interpretation of density trends across sites and years. The incomplete sampling of sites across all years reduced the ability to make clear comparisons among sites and years, in part because there was only one year in which all three sites were sampled. Further analysis of these data to assess trends over time within each site should be considered, especially as future sampling efforts are implemented. Changes to the spatial targeting of transects in 2017 (see ROV Methods Appendix) should also be kept in consideration when planning and analyzing future sampling efforts.

We are able to detect natural, interannual variability in density for select species with ROV sampling.

Even though ROV sampling occurs at infrequent intervals, there were a number of species-specific, inter-annual patterns detected. Among the species exhibiting inter-annual changes in density, most showed increases, with only two significant declines. A large recruitment event for Red Urchins was detected. Changes in Sea Star densities provide a more comprehensive picture of subtidal Sea Star Wasting Syndrome at deeper reef locations. This report highlight the dramatic change in Sunflower Stars, which were essentially eradicated. The ROV Sea Star Wasting Disease Report highlights the impacts of the disease on other sea stars at Cascade Head and its associated comparison areas. Short Red Gorgonians declined substantially between 2012 and 2017 for unknown reasons.

Despite a wealth of information from ROV monitoring surveys, the continuity of future sampling is uncertain without any increase in support

From an ROV monitoring perspective, sampling occurs at irregular intervals because of the high cost of chartering vessels for ROV sampling and the small budget of the Marine Reserves Program. Much of the sampling that was conducted between 2012 and 2017 was enabled by successfully pursuing external funding for various research topics and capitalizing on the funding to conduct the research at Marine Reserve and comparison area sites. A federal grant provided funding for charter costs in surveys prior to 2017, but that grant source became unavailable from that point onward. The ODFW Marine Reserves and Habitat programs have struggled with reporting results of monitoring data at regular intervals because of the small budget and staff of both programs. Despite these challenges, a wealth of information lies in the data gathered from ROV monitoring, including the ability to understand species-habitat relationships in both the marine reserve and its two comparison areas. Importantly, the program has accumulated significant methodological and analytical infrastructure (e.g. well-developed protocols, databases, video review skill, statistical and interpretive skill, and computer code) that can facilitate much more efficient cycles of data collection and reporting in the future. The Marine Reserves program will attempt to continue data collection at the Cascade Head Marine Reserve and surrounding comparison areas with the ROV, ideally if a new grant source could be identified or its base budget increased. Without new funds, continued sampling, even at irregular intervals, as well as analysis and reporting will continue to be a challenge for the program.

3 ROV Invertebrate Methods

Detailed methods documenting the survey design, field sampling methods, and video review methods for the Remotely Operated Vehicle (ROV) video sampling are presented in the ROV Methods Appendix. The following sections briefly summarize the ROV methods and describe the data treatment and analytical approach for invertebrates for this synthesis.

3.1 ROV video sampling

Remotely Operated Vehicle (ROV) video sampling is conducted in the Cascade Head Marine Reserve, Cavalier Comparison Area and Schooner Creek Comparison Area. Monitoring began in 2012, and occurred at irregular intervals because of the high cost of chartering vessels for ROV sampling and the small budget of the Marine Reserves Program. Prioritization of comparison areas underwent some changes during the first years of sampling, and as a consequence not all sites were sampled in every year. Sampling occurred once in each sampled year - either in spring or fall, with variable effort across years depending on the availability of external funds to support vessel charters. In 2012 and 2013, sampling was conducted in the fall (Sept-Oct), while in 2017 sampling was conducted in the spring (May). Each day, approximately 8-14 500 m long transects were surveyed from a list of randomized transects that intersect a minimum proportion of mapped rocky substrate at the appropriate depths (see ROV Methods Appendix for more detail on transect selection and sampling protocols).

All video data collected from ROV sampling were reviewed and filtered to meet data quality criteria before inclusion in analysis. Segments of transects with poor visibility, terrain obstructions, or piloting actions that invalidate the assumptions of belt transect sampling were excluded. Transect widths were derived through measurement of the on-screen width of a pair of parallel lasers. Along-transect distance was derived from an acoustic ROV tracking system. Transect length and width were multiplied to calculate the total area viewed, forming the denominator for organism density calculations. Observed invertebrates belonging to a select list of taxa targeted for video review (see ROV methods appendix) were identified to species where possible, and otherwise were recorded in higher level taxonomic groupings. Primary and secondary substrates were assessed continuously along the transect.

Due to the high density of Short Red Gorgonians at some sites, a specialized video review technique was used to subsample the available videos and count gorgonians intercepted by a restricted portion of the full video review frame. Densities were then extrapolated to full-frame equivalents using ratios derived from pilot studies that counted gorgonians with both the full-frame and reduced-frame protocols. Density data for Short Red Gorgonians were not available for 2013.

3.2 Data filtering

To account for varying suitability of portions of video for assessing the density of small invertebrates, data were filtered according to a categorization called ViewScale (see ROV Methods Appendix). This limited the use of video data to discrete segments along transects that provided consistent ability to detect macroinvertebrates such as seastars and sea urchins, categorized as ViewScale = 3.

Substrate types were classified during video review according to primary and secondary habitats continuously along each transect (see ROV Methods appendix). The various observed combinations of primary and secondary substrate types were reduced for this analysis to two overall categories: “soft substrates”, composed of substrate groups sand and gravel, and “hard substrates”, composed of substrate groups cobble, boulder, and bedrock.

For the invertebrate analyses, we excluded all soft substrate data in order to compare similar hard-substrate habitats across sites and years. Preliminary assessment of the soft substrate associated invertebrate data showed that differences among transects in the proportion of soft substrate habitat sampled had the potential to skew the perception of differences in invertebrate densities and community composition among sites and years. Invertebrate densities in soft substrates were plotted in comparison to hard-substrate densities, and the plots (not included in this report) show very few observations of hard-substrate-associated species (e.g. rock scallops) in transect portions tagged as soft substrate, confirming the utility of the substrate data for filtering out soft-substrate portions of transects.

To explore invertebrate relationships with specific substrate types within the overall hard substrate category, we included the percent of each transect’s data that was categorized as boulder as a potential covariate in the analyses. Boulders are of special interest because of their ability to provide distinct habitats such as large protected interstitial spaces.

3.3 Data aggregation and sample units

Invertebrate densities were calculated and analyzed at the transect scale. Invertebrate counts and total viewed area were summed across all ViewScale 3, hard-substrate portions of transects, so that transects are the sample unit for invertebrate analysis. Aggregating the discrete segments of ViewScale 3 data across each transect helps reduce the effect of any inaccuracies in the transect view area estimate, because summing the total view area across the transect helps average out over- and under-estimated view areas for specific points along the transect.

For additional details on data collection, video review and data filtering, please review documentation in the ROV Methods Appendix.

3.4 Data Analysis

3.4.1 Diversity

With ROV invert surveys, we explored several concepts related to species diversity at a given site:

  • species richness
  • unique, common & rare species
  • diversity indices
  • diversity through time

3.4.1.1 Species Richness

To explore species richness at a given site, we reported total observed species richness and also calculated total estimated species richness.

We report total observed species richness at a given site using incidence data across all sampling years because each site (reserve or comparison area) likely has a species pool larger than can be sampled in any one year. We excluded unidentified species from the summaries. Species richness metrics are highly sensitive to survey effort. While the ROV survey targets transect lengths of 500m, habitat and ocean conditions result in highly variable transect lengths. In order to overcome the confounding factor of transect size on species richness, diversity rarefaction curves were standardized by the number of individuals observed (Gotelli and Colwell 2001; Chao et al 2014).

To calculate estimated total species richness, we used a rarefaction and extrapolation technique as described in Hsieh et al 2016, to calculate the effective number of species at each given site. This is the equivalent of calculating Hill diversity = 0. Hill numbers represent a unified standardization method for quantifying and comparing species diversity across multiple sites (Hill 1973), and they represent an intuitive and statistically rigorous alternative to other diversity indices (Chao et al 2014).

We used individual-based abundance data and the iNext package in R to estimate the asymptote of the species accumulation curve, or the estimated total number of species observable by ROV at a given site. These curves are expressed as the mean expected number of species per number of individuals observed. We also calculated confidence intervals associated with these rarefaction and extrapolation curves and can therefore compare across sites to explore similarity of total estimated species richness for a given sampling effort.

3.4.1.2 Unique, Common, and Rare Species

Richness alone does not sufficiently describe species biodiversity; additionally uniqueness, rarity and common species also shape and define concepts of biodiversity.

As a first step to exploring unique, rare and common species we generated species count tables at the transect level. These tables exclude the unidentified individuals and species not well targeted by the ROV. The species count tables include a total count for each species summed for all years by site, and for each year-site combination, as well as mean frequency of occurrence across all samples. This information can tell us both about how frequently the species is observed, as well as its relative abundance.

Frequency of occurrence is defined here as the proportion of surveys that contained a given species. From the species count tables we identified rare species, as those with a frequency of occurrence of 10% or less (Green and Young 1993), and common species as those with a frequency of occurrence greater than 50% (in other words, the species is observed one out of every two transects). We also identified species that were unique to each marine reserve and comparison area.

3.4.1.3 Diversity Indices

To gain additional insight into species diversity, we explored several diversity indices by comparing Hill diversity numbers across sites using the iNEXT diversity package in R (Hsieh et al 2016). Hill numbers are parameterized by a diversity order q, which determines the measures’ sensitivity to species relative abundances (Hsieh et al 2016). Hill numbers include the three most widely used species diversity measures; species richness (q = 0), Shannon diversity (q = 1) and Simpson diversity (q = 2) (Hsieh et al 2016). We used individual based abundance data with the iNext package in R, to plot rarefaction and extrapolation curves for each Hill number, and compare results across sites. We also calculated 95% confidence intervals associated with these rarefaction & extrapolation curves.

3.4.1.4 Diversity Through Time

Finally we explored how diversity changed through time. First we plotted each species yearly rarefaction curve against the total cumulative rarefaction curve for all years combined to determine if we had sampled appropriately to compare species diversity from year to year.

All analyses and graphs were created in R v4.0.2, using the iNEXT and Vegan packages. In some cases, the common names for species displayed in tables and graphics in the diversity section of the report differ from those used elsewhere. For clarity, the common pairs are pseudonyms: California Sea Cucumber = Giant Sea Cucumber; Giant Plumose Anemone = White Plumose Anemone; Purple Hinged Rock Scallop = Rock Scallop;

3.4.2 Community Composition

We focused our community composition analysis on the question of whether variation in density was driven by spatial (site) or temporal (year) factors. We did this through both data visualizations with non-multidimensional scaling (nMDS) plots and with statistical tests such as principal coordinates analyses (PCO), multivariate ANOVA tests (PERMANOVA), and dispersion tests (PERMDISP). In addition to site and year, we also explored depth and proportion of boulder habitat as potential habitat-related drivers of variation.

To explore variation by site and year, we used transformed (using Wisconsin double standardization) invertebrate density data calculated from ROV count data (# individuals / area) so a similarity-based resemblance matrix was selected. Short Red Gorgonian data were not available for 2013, so this species was excluded from all community composition analyses. To visualize the clustering or spread of the multivariate dataset with respect to the key variables we plotted 2D nMDS biplots symbolized by site and year.

To test the statistical significance of variation by site and year we ran a permutational analysis of variance (PERMANOVA), using Site and Year as fixed factors and Depth and Percent Boulder as continuous covariates. To explore if any significant results of the PERMANOVA were related to differences in location or differences in dispersion of samples (among sites or among years), we ran a permutational dispersion test, a distance based test for homogeneity of multivariate dispersions (Anderson and Walsh 2013). Significant heterogeneity of dispersions can lead to erroneously significant PERMANOVA results, so this test is used to distinguish where differences in dispersion may be influential in interpreting PERMANOVA results.

To better understand the quantitative contribution of various factors in explaining variation in the data, we ran a principal coordinates (PCO) analysis using a Bray-Curtis resemblance matrix, providing information on the percent of variation explained by each axis. To identify the species most strongly correlated with the PCO ordination, we used a vector analysis and displayed species vectors on the PCO plot for those species with significant correlations and r^2 > 0.35, a threshold selected on the basis of yielding a manageably small number of species for presentation. We also plotted individual species bubble plots showing the density of the highly correlated species on the PCO ordination to visualize their abundance relative to the two PCO axes.

To explore the relationship of environmental variables with the observed patterns in community structure, we also used the PCO ordination to display trends in depth, season and percent cover of boulder habitat.

The community composition analyses (NMDS, PCO, PERMANOVA, and dispersion tests) were implemented in R using package “vegan” v. 2.5-7 (Oksanen et al 2020).

3.4.3 Abundance

Invertebrate density data were generated per transect by summing individual counts across the ViewScale 3, hard-substrate portions of each transect and dividing by the view area summed across those same portions. Densities are expressed in plots as individuals per 100 square meters, a convention used simply for better readability of typically low-density figures. 95% confidence intervals of density were generated as +/- 1.96 times the standard error. The plots display unweighted means and confidence intervals; preliminary explorations used weighted means and standard errors that weighted transects by their total viewed area in order to account for variation in effective sampling effort among transects, but these plots showed very minimal differences and are not included here. Individual species density data are presented for the selected focal species and additionally for species identified as influential in community composition analyses.

3.4.3.1 Aggregate abundance

Densities of invertebrate species within several major taxonomic groupings were generated following the same procedures. Groupings were as follows: Anemone, Barnacle (mostly Giant Barnacle), Bivalve (mostly Rock Scallop), Cephalopod (Red Octopus and Giant Pacific Octopus), Crustacean (including Dungeness Crab), Cucumber (including Burrowing Sea Cucumber and California Sea Cucumber), Nudibranch, Ophiuroid (e.g. brittle stars, basket stars), Sea Star, Sponge (colonies counted as individuals), Tunicate, Urchin (mostly Red Urchin, also Purple Urchin). Densities of aggregate groups are presented graphically but were not statistically analyzed. Several other taxa that were observed are not summarized because they were not consistently reviewed by the same methodology across all years, including polychaetes, anthozoans, and gastropods.

3.4.3.2 Individual species abundance

Statistical analyses of invertebrate density were conducted on the selected focal species and a few additional species that were highlighted as important in community composition analyses. Data explorations suggested the potential for influential nonlinear relationships of invertebrate densities with continuous covariates Depth and Percent Boulder. We employed generalized additive models (GAMs) incorporating smooth functions of the covariates along with fixed-effects factors Site and Year to compare individual invertebrate species densities across sites and years. We first developed a fixed set of possible GAMs, and selected the model with the lowest AIC. When two models AIC scores were effectively tied, we chose the simpler model. All models included Site and Year. Four models additionally contained linear effects of either Depth or Percent Boulder, both covariates, or neither covariate. The remainder of the set included models with and without all combinations of the following:

  • with or without a Site * Year interaction
  • a single Depth smooth across all sites, or a separate Depth smooth for each site
  • a linear Percent Boulder covariate across all sites, or a single Percent Boulder smooth across all sites, or a separate Percent Boulder smooth for each site

For all potential model covariates, the smoothness parameter ‘k’ was fixed at 3 in order to avoid overfitting. The selected model varied among species, and is reported in the results section for each species. The GAM modeled count of individuals using a negative binomial distribution with a log link, with viewed area included as an offset, thereby accounting for density across transects with varying total survey areas. Therefore the model coefficients provided in tables are in log space (i.e. exponentiating the coefficient estimates puts them on the density scale of individuals per square meter). Smooth plots of covariates are provided showing the predicted invertebrate density across the range of each continuous covariate while the other factors and covariates in the model are held at fixed (mean) values. Where the selected model included separate smooth functions for each site, smooth plots at all three sites are displayed in the results section. Where the selected model included a single smooth across sites, just a single plot is displayed. In some smooth plots the extent of the shaded 95% confidence interval is truncated by limiting the display of the y axis, allowing better visualization of patterns in the predicted mean density. This generally happened when model fits were poor and does not affect interpretation of significant covariate smooths.

In certain cases, absence of species precluded analysis across all years. In this situation, models were run on individual years’ data. In other cases, certain species were not sufficiently abundant at a given site to warrant inclusion in the statistical analyses. In these cases, the site in question was dropped from the analysis if the data violated the required assumptions.

Where significant Site * Year interactions were identified, we did not pursue further lower-level analyses comparing individual years or sites. Rather we restricted our inferences to the significance of each factor level identified in the GAM’s summary table, which treated the Cascade Head Marine Reserve as the reference level of Site against which the other sites were compared, and 2012 as the reference year against which other years were compared. All analyses were conducted in R (R Core Team (2021)). GAM models of individual species densities were implemented in package “mgcv” v. 1.8-35 (Wood 2011).

\(~\)

\(~\)

4 Cascade Head Results

4.1 Sampling effort summary

ROV sampling efforts at Cascade Head and its comparison areas resulted in three years of data collection, where varying sample sizes were collected per year (Fig. 2). The last year of sampling (2017) resulted in the largest sampling effort, and the only one that surveyed all sites. Schooner Creek Comparison Area was not surveyed in 2012 and Cavalier Comparison Area was not surveyed in 2013.

Fig. 2: ROV monitoring efforts represented as number of transects at the Cascade Head Marine Reserve and surrounding comparison areas.

Fig. 2: ROV monitoring efforts represented as number of transects at the Cascade Head Marine Reserve and surrounding comparison areas.

4.1.1 Sampled depth distribution

Differences in the depth distribution of sampling at the Cascade Head Marine Reserve relative to the two comparison areas may be important in interpreting species’ density patterns.

Figure DPTH presents the total survey area included in invertebrate density analyses in each Year and Site within 5 m depth intervals. The distribution of available seafloor depths varies among the Cascade Head Marine reserve, which has a larger proportion of shallower area, and the two comparison areas, which have proportionally more deep areas. Compounding the depth disparities in sampling, numerous environmental and logistical factors affected the ability of the ROV to acquire video data that would ultimately pass all data quality and habitat-based filtering steps and be included in analyses. Daily ROV operations were sometimes limited by water clarity or currents, which both tended to vary across depths within sites. In these cases, the field crew generally substituted other randomly selected transects in areas (depths or sites) that were productive for sampling. Therefore ROV sampling resulted in varying degrees of effort across depths. Finally, the distinct geomorphology of each Site (i.e. the amount of rocky and sandy habitats at different depths) influenced the amount of data that was excluded by the filtering step that excluded all soft-substrate portions of transects.

Fig. 3: Depth distribution of total survey area included in fish density analyses at the Cascade Head Marine Reserve and its associated comparison areas. The area includes only portions of transects that passed all data quality and habitat-based filtering steps for inclusion in invertebrate density analysis. X-axis labels indicate the shallower end of each 5 m depth interval.

Fig. 3: Depth distribution of total survey area included in fish density analyses at the Cascade Head Marine Reserve and its associated comparison areas. The area includes only portions of transects that passed all data quality and habitat-based filtering steps for inclusion in invertebrate density analysis. X-axis labels indicate the shallower end of each 5 m depth interval.

\(~\)

\(~\)

4.2 Diversity

4.2.1 Species richness

Species richness at Cascade Head Marine Reserve is very similar to the comparison areas

Over the four years of sampling with the ROV a total of 39 species (or species groups) were observed in the Cascade Head Marine Reserve (Table 3). The Schooner Creek Comparison Area (n = 42) and Cavalier Comparison Area (n = 40) had similar numbers of invert species observed (Table 3). These observed numbers of species richness are similar to the estimated numbers of total species richness (Table 3).

Table 3: Observed and estimated species richness by site with lower (LCL) and upper (UCL) 95% confidence limits
Area Observed_Richness Estimated_Richness LCL UCL
Cascade Head Marine Reserve 39 39 39 40
Schooner Creek Comparison Area 42 43 42 50
Cavalier Comparison Area 40 41 40 54

\(~\)

\(~\)

Species rarefaction curves highlight that at small samples sizes , such as those for any given year, the species richness among sites is very similar (Fig. 4). Rarefaction curves levels off, suggesting saturation in species richness with this tool at this site.

Fig. 4: Species rarefaction curves for the Cascade Head Marine Reserve and its two comparison areas. Data are pooled across all years of sampling for each site.

Fig. 4: Species rarefaction curves for the Cascade Head Marine Reserve and its two comparison areas. Data are pooled across all years of sampling for each site.

\(~\)

\(~\)

4.2.2 Unique, common and rare species

The number of common and rare species differs among the survey areas; however, the number of unique species is similar across all sites.

One unique species were observed at the Cascade Head Marine Reserve: the Tube Anemone. At Schooner Creek Comparison Area, Pink Tipped Green Anemone, Purple Urchin, and the Vermilion Star were unique. There were no unique species at Cavalier Comparison Area.

The Cascade Head Marine Reserve (n = 14) had a moderate number of common species compared with Schooner Creek Comparison Area (n = 20) and the Cavalier Comparison Area (n = 10). Out of the 14 common species identified at Cascade Head Marine Reserve, 11 of those were shared with Schooner Creek Comparison Area and 9 with Cavalier Comparison Area. The greatest number of rare species were observed at Cavalier Comparison Area (n = 11) followed by Cascade Head Marine Reserve (n = 9) and Schooner Creek Comparison Area (n = 7).

Not all species were observed each year, for a summary of species counts over the years by site please see tables below.

Pooled species counts across all years and species counts by individual sampling year are included in the following tables:

4.2.2.1 Cascade Head Marine Reserve

Fig. 5: Relative frequency of occurrence of species observed at the Cascade Head Marine Reserve and its associated Comparison Areas in ROV transects. See separate tabs for each site.

Fig. 5: Relative frequency of occurrence of species observed at the Cascade Head Marine Reserve and its associated Comparison Areas in ROV transects. See separate tabs for each site.

4.2.2.2 Schooner Creek Comparison Area

Fig. 5: Relative frequency of species observed at the Cascade Head Marine Reserve and the comparison areas in ROV transects. See separate tabs for each site.

Fig. 5: Relative frequency of species observed at the Cascade Head Marine Reserve and the comparison areas in ROV transects. See separate tabs for each site.

4.2.2.3 Cavalier Comparison Area

Fig. 5: Relative frequency of species observed at the Cascade Head Marine Reserve and the comparison areas in ROV transects. See separate tabs for each site.

Fig. 5: Relative frequency of species observed at the Cascade Head Marine Reserve and the comparison areas in ROV transects. See separate tabs for each site.

\(~\)

\(~\)

4.2.3 Diversity Indices

Slight differences in the fish community among the marine reserve, Schooner Creek Comparison Area and Cavalier can be seen when comparing rarefaction and interpolation of the hill diversity numbers across the three sites (Fig. 6). When q = 0, this is the equivalent of comparing species richness across sites, species richness is nearly equivalent with the greatest species richness in Schooner Creek Comparison Area. The rarefaction curves reach an asymptote across all three survey sites.

When q = 1, a derivation of the Shannon-Weiner diversity index across sites, there is a greater number of effective species at the Cascade Head Marine Reserve compared to the the other two comparison areas (Fig. 6).

When q = 2, a derivation of the Simpson evenness index across sites, the Cascade Head Marine Reserve also has a greater evenness of abundance of species than either comparison area (Fig. 6).

The raw count of total individuals observed was: Cascade Head Marine Reserve (n = 28898), Schooner Creek Comparison Area (n = 33412) and Cavalier Comparison Area (n = 23579).

Fig. 6: Comparing Hill diversity numbers across the Cascade Head Marine Reserve and its associated Comparison Areas from ROV samples. Hill numbers include the three most widely used species diversity measures; species richness (q = 0), Shannon diversity (q=1) and Simpson diversity (q=2) (Hsieh et al 2016). Note sample unit is an individual organismFig. 6: Comparing Hill diversity numbers across the Cascade Head Marine Reserve and its associated Comparison Areas from ROV samples. Hill numbers include the three most widely used species diversity measures; species richness (q = 0), Shannon diversity (q=1) and Simpson diversity (q=2) (Hsieh et al 2016). Note sample unit is an individual organism

Fig. 6: Comparing Hill diversity numbers across the Cascade Head Marine Reserve and its associated Comparison Areas from ROV samples. Hill numbers include the three most widely used species diversity measures; species richness (q = 0), Shannon diversity (q=1) and Simpson diversity (q=2) (Hsieh et al 2016). Note sample unit is an individual organism

\(~\)

\(~\)

4.2.4 Diversity through time

We did not sample enough to evaluate meaningful annual changes in species diversity at the Cascade Head Marine Reserve and its comparison areas

While annual species rarefaction curves indicate asymptotic estimates of species richness, there were no discernible inter-annual patterns. Both Cavalier Comparison Area and Schooner Creek Comparison Area were only sampled for two years, precluding meaningful temporal analysis of invert species richness. Slight differences in total species obtained are likely driven by the presence or absence of rare species during the ROV survey or by differences in relative habitat type surveyed.

Fig. 7: Species rarefaction curves by year and Area from ROV data. Note that x-axis is scaled by 1000s of individuals.

Fig. 7: Species rarefaction curves by year and Area from ROV data. Note that x-axis is scaled by 1000s of individuals.

\(~\)

\(~\)

4.3 Community Composition

4.3.1 Variation by Site and Year

Differences in the invertebrate community structure among the Cascade Head Marine Reserve and its associated comparison areas were minor.

There was little apparent structure in the 2D nMDS plot relative to Site (Fig. 8). The relatively poor ability of the multi-dimensional ordination to be represented in two dimensions was indicated by the high stress level of 0.2, a value that suggests interpretation should be restricted to major, obvious features of the plot. The PERMANOVA results indicated significant effects of Site and Year, in addition to a Site * Year interaction and significant effects of the covariates Depth and Percent Boulder (p < 0.05, Table 10). Despite statistical significance, the proportion of variation explained by these factors was relatively low; Site explained only 8% of the variation, and the significant interaction means that Site was not an important predictor of community structure in all years.

Differences in community composition by year were apparent, accounting for 18% of the total variation in community composition.

Among the three years of sampling, transects from 2012 and 2017 formed the most distinct groupings in the nMDS plot (Fig. 8), with the other year 2013 intermediate and overlapping to a greater degree with 2012. This observation is consistent with a progression over time in the composition of the invertebrate community, rather than an effect of the year-to-year changes in sampling of the three sites. The PERMANOVA analysis showed that Year explained 18% of the variation in community composition (p < 0.05, Table 10).

Heterogeneity in dispersions within sites and years contributed to significant PERMANOVA results.

Assessment of the group dispersions showed significant heterogeneity of dispersions among years and among sites (p < 0.05, Table 11, Table 12). Pairwise comparison of individual years and sites by the Tukey HSD test showed that the Cavalier Comparison Area had greater dispersion than the other two sites (Table 13), and that dispersion was lower in 2013 than in 2012 and 2017 (Table 14). These results indicate that the significant PERMANOVA result described above is likely influenced by differences among sites and years in the inherent variability in community structure among transects (a “dispersion effect”) in addition to any contribution of systematic differences in the structure of communities (a “location effect”).

4.3.1.1 NMDS plot by Site

Fig. 8: Results from nMDS plots with ROV invertebrate data, demonstrating similarity in invertebrate community composition at the Cascade Head Marine Reserve and its surrounding comparison areas. The raw density data were transformed and a Bray-Curtis similarity matrix was used. See separate tabs for site and year.

Fig. 8: Results from nMDS plots with ROV invertebrate data, demonstrating similarity in invertebrate community composition at the Cascade Head Marine Reserve and its surrounding comparison areas. The raw density data were transformed and a Bray-Curtis similarity matrix was used. See separate tabs for site and year.

4.3.1.2 NMDS plot by Year

Fig. 8: Results from nMDS plots with ROV invertebrate data, demonstrating similarity in invertebrate community composition at the Cascade Head Marine Reserve and its surrounding comparison areas. The raw density data were transformed and a Bray-Curtis similarity matrix was used. See separate tabs for site and year.

Fig. 8: Results from nMDS plots with ROV invertebrate data, demonstrating similarity in invertebrate community composition at the Cascade Head Marine Reserve and its surrounding comparison areas. The raw density data were transformed and a Bray-Curtis similarity matrix was used. See separate tabs for site and year.

\(~\)

\(~\)

\(~\)

\(~\)

4.4 Aggregate Density

Trends in aggregate density of various taxonomic categories are presented in Figure AGG. The most notable trends in aggregate density across the sampling period were the overall declines in Sea Stars (associated with the Sea Star Wasting Syndrome that affected sea stars across the west coast) and in sponges, and the increase in Sea Cucumbers (both Burrowing Cucumbers and California Sea Cucumbers) at two of the three sites. Other trends in aggregate abundance are actually driven by a single member of the group, including Ophiuroids (reflecting Basket Stars) and Urchins (reflecting Red Sea Urchins).

Fig. 11: Aggregate invertebrate density by taxonomic category with 95% confidence intervals at the Cascade Head Marine Reserve and its associated comparison areas.

Fig. 11: Aggregate invertebrate density by taxonomic category with 95% confidence intervals at the Cascade Head Marine Reserve and its associated comparison areas.

\(~\)

\(~\)

4.5 Focal Species Density

4.5.1 Ochre Seastar (Pisaster ochraceus)

Ochre Seastars are a shallow-dwelling species not suited for sampling with the ROV

Ochre Seastars were observed by the ROV only occasionally at the shallowest depths, generally shallower than the depth range targeted for ROV sampling. Accordingly no statistical analysis of their densities was conducted. Overall densities are shown in Fig. 12.

Fig. 12: Ochre Star density with 95% confidence intervals at the Cascade Head Marine Reserve and its associated comparison areas.

Fig. 12: Ochre Star density with 95% confidence intervals at the Cascade Head Marine Reserve and its associated comparison areas.

\(~\)

\(~\)

4.5.2 Sunflower Star (Pycnopodia helianthoides)

Complete disappearance of the Sunflower Star across all sites

The Sunflower Star was eradicated by the Seastar Wasting Syndrome that devastated seastar communities across the West Coast starting in 2013 and 2014. Total counts were reduced to zero in 2017 (Table 15).

Because the loss of most Sunflower Stars by 2017 complicates any statistical modeling of the full dataset, we restricted this assessment to 2012 and 2013 data. In 2012, the Cascade Head Marine Reserve had a lower density of Sunflower Stars than the Cavalier Comparison Area (p < 0.05, Fig. 13, Table 17). The selected GAM model included individual smooth effects of depth at each site and a single linear effect of percent boulder. The selected model was:

Count = Year + Site + s(Depth, by = Site, k = 3) + Boulder.pct, offset = log(area), family = nb

GAM model results can be found in the links below:

Sunflower Star density had a unimodal relationship with depth at the Cavalier Comparison Area, peaking at around 30 m (p < 0.05, Fig. 13, Table 18), but no significant relationship with depth at the other two sites. The relationship between Sunflower Star density and percent boulder was non-signicant (Table 17).

4.5.2.1 Sunflower Star Density Timeseries

Fig. 13: Sunflower Star density with 95% confidence intervals at the Cascade Head Marine Reserve and its associated comparison areas. See separate tabs for density timeseries and plots of GAM covariate smooths.

Fig. 13: Sunflower Star density with 95% confidence intervals at the Cascade Head Marine Reserve and its associated comparison areas. See separate tabs for density timeseries and plots of GAM covariate smooths.

4.5.2.2 GAM smooth for Depth

Fig. 13: Sunflower Star GAM smooth for depth at the Cascade Head Marine Reserve and its associated comparison areas. The value is the predicted density (+/- 95% confidence interval) across the range of depths, holding other factors in the model constant.

Fig. 13: Sunflower Star GAM smooth for depth at the Cascade Head Marine Reserve and its associated comparison areas. The value is the predicted density (+/- 95% confidence interval) across the range of depths, holding other factors in the model constant.

\(~\)

\(~\)

4.5.3 Purple Sea Urchin (Strongylocentrotus purpuratus)

Purple Sea Urchins, a shallow-dwelling species, were mostly absent at the depths surveyed by the ROV.

Purple Sea Urchins are generally shallower than the depth range targeted for ROV sampling and are not well suited for ROV sampling. A total of 3 Purple Sea Urchins was observed by the ROV, all in 2017 at the Cavalier Comparison Area.

\(~\)

\(~\)

4.5.4 Red Sea Urchin (Mesocentrotus franciscanus)

Recruitment of Red Sea Urchins at the Cascade Head Marine Reserve and the Schooner Creek Comparison Area

Red Sea Urchin density increased substantially at the Cascade Head Marine Reserve and the the Schooner Creek Comparison Area across the study period (Fig. 14). Only two Red Sea Urchins were seen in each of 2012 and 2013, so data analysis was restricted to data from 2017. Overall Red Sea Urchin density at the Cascade Head Marine Reserve in 2017 was higher than at the Cavalier Comparison Area and similar to that at the Schooner Creek Comparison Area (p < 0.05). With Year removed from the analysis, the selected GAM model included only Site and a linear effect of percent hard boulder. The selected model was:

Count = Site + Boulder.pct, offset = log(area), family = nb

GAM model results can be found in the links below:

Red Sea Urchin density in 2017 did not vary significantly with depth, but had a decreasing linear response to an increasing proportion of boulder substrate (p < 0.05, Fig. 14).

4.5.4.1 Red Sea Urchin Density Timeseries

Fig. 14: Red Sea Urchin density with 95% confidence intervals at the Cascade Head Marine Reserve and its associated comparison areas.

Fig. 14: Red Sea Urchin density with 95% confidence intervals at the Cascade Head Marine Reserve and its associated comparison areas.

\(~\)

\(~\)

4.5.5 Rock Scallop (Crassadoma gigantea)

Higher Rock Scallop densities at the Cascade Head Marine Reserve than at the Cavalier Comparison Area.

The Cascade Head Marine Reserve had higher Rock Scallop densities than the Cavalier Comparison Area across years (p < 0.05, Fig. 15, Table 21, Table 22).

Increasing Rock Scallop densities across the sampling period.

Overall Rock Scallop density increased between 2012 and 2017 (p < 0.05, Fig. 15, Table 21, Table 22). The selected GAM model excluded the Site * Year interaction and included individual smooth effects of depth at each site and a single linear effect of percent boulder. The selected model was:

Count = Year + Site + s(Depth, by = Site, k = 3) + Boulder.pct, offset = log(area), family = nb

GAM model results can be found in the links below:

Rock Scallop density had a unimodal relationship with depth at the Cavalier Comparison Area, peaking at around 30 m depth (Fig. 15, Table 23), but the decreasing trend with depth at the other two sites was not significant. The relationship with percent boulder substrate was not significant.

4.5.5.1 Rock Scallop Density Timeseries

Fig. 15: Rock Scallop density with 95% confidence intervals at the Cascade Head Marine Reserve and its associated comparison areas. Data are See separate tabs for density timeseries and plots of GAM covariate smooths.

Fig. 15: Rock Scallop density with 95% confidence intervals at the Cascade Head Marine Reserve and its associated comparison areas. Data are See separate tabs for density timeseries and plots of GAM covariate smooths.

4.5.5.2 GAM smooth for Depth

Fig. 15: Rock Scallop GAM smooth for depth at the Cascade Head Marine Reserve and its associated comparison areas. The value is the predicted density (+/- 95% confidence interval) across the range of depths, holding other factors in the model constant.

Fig. 15: Rock Scallop GAM smooth for depth at the Cascade Head Marine Reserve and its associated comparison areas. The value is the predicted density (+/- 95% confidence interval) across the range of depths, holding other factors in the model constant.

\(~\)

\(~\)

4.5.6 California Sea Cucumber (Parastichopus californicus)

Higher densities of California Sea Cucumbers at the Cascade Head Marine Reserve than at either comparison area.

California Sea Cucumber densities were higher at the Cascade Head Marine Reserve than at either comparison area (p < 0.05, Fig. 16, Table 24, Table 25).

Increasing densities of California Sea Cucumbers across the study period.

California Sea Cucumber density increased between 2012 and 2017 (p < 0.05, Table 25). The selected GAM model excluded the Site * Year interaction and included a single smooth effect of depth across sites, and a single linear effect of percent boulder. The selected model was:

Count = Year + Site + s(Depth, k = 3) + Boulder.pct, offset = log(area), family = nb

GAM model results can be found in the links below:

California Sea Cucumber density across sites had a nonlinear, increasing relationship with depth (p < 0.05, Fig. 16, Table 26), reaching a peak density at around 35 m depth. The relationship with percent boulder was nonsignificant.

4.5.6.1 California Sea Cucumber Density Timeseries

Fig. 16: California Sea Cucumber density with 95% confidence intervals at the Cascade Head Marine Reserve and its associated comparison areas. Data are See separate tabs for density timeseries and plots of GAM covariate smooths.

Fig. 16: California Sea Cucumber density with 95% confidence intervals at the Cascade Head Marine Reserve and its associated comparison areas. Data are See separate tabs for density timeseries and plots of GAM covariate smooths.

4.5.6.2 GAM smooth for Depth

Fig. 16: California Sea Cucumber GAM smooth for depth at the Cascade Head Marine Reserve and its associated comparison areas. The value is the predicted density (+/- 95% confidence interval) across the range of depths, holding other factors in the model constant.

Fig. 16: California Sea Cucumber GAM smooth for depth at the Cascade Head Marine Reserve and its associated comparison areas. The value is the predicted density (+/- 95% confidence interval) across the range of depths, holding other factors in the model constant.

\(~\)

\(~\)

4.5.7 White Plumose Anemone (Metridium farcimen)

Highly variable densities of White Plumose Anemones among transects, but no consistent difference among sites or years.

White Plumose Anemone densities were characterized by high variability among transects, especially in 2012, contributing to an overall conclusion of no differences between sites and years (Fig. 17, Table 28). There was a trend toward higher densities at the Cavalier Comparison Area than at the Cascade Head Marine Reserve, but the overall effect of Site in the model was marginally non-significant (Table 27, Table 28). The selected GAM model included a single smooth effect of depth across sites, and individual smooths of percent boulder for each Site. The selected model was:

Count = Year + Site + s(Depth, k = 3) + s(Boulder.pct, by = Site, k = 3), offset = log(area), family = nb

GAM model results can be found in the links below:

There was no significant response of White Plumose Anemone density to depth (Fig. 17, Table 29). White Plumose Anemone density decreased nonlinearly with percent boulder at the Cascade Head Marine Reserve, and had a unimodal relationship at the Cavalier Comparison Area (p < 0.05, Fig. 17, Table 29), but no significant response at the Schooner Creek Comparison Area.

4.5.7.1 White Plumose Anemone Density Timeseries

Fig. 17: White Plumose Anemone density with 95% confidence intervals at the Cascade Head Marine Reserve and its associated comparison areas. Data are See separate tabs for density timeseries and plots of GAM covariate smooths.

Fig. 17: White Plumose Anemone density with 95% confidence intervals at the Cascade Head Marine Reserve and its associated comparison areas. Data are See separate tabs for density timeseries and plots of GAM covariate smooths.

4.5.7.2 GAM smooth for Depth

Fig. 17: White Plumose Anemone GAM smooth for depth at the Cascade Head Marine Reserve and its associated comparison areas. The value is the predicted density (+/- 95% confidence interval) across the range of depths, holding other factors in the model constant.

Fig. 17: White Plumose Anemone GAM smooth for depth at the Cascade Head Marine Reserve and its associated comparison areas. The value is the predicted density (+/- 95% confidence interval) across the range of depths, holding other factors in the model constant.

4.5.7.3 GAM smooth for boulder

Fig. 17: White Plumose Anemone GAM smooth for percent boulder at the Cascade Head Marine Reserve and its associated comparison areas. The value is the predicted density (+/- 95% confidence interval) across the range of percent boulder, holding other factors in the model constant.

Fig. 17: White Plumose Anemone GAM smooth for percent boulder at the Cascade Head Marine Reserve and its associated comparison areas. The value is the predicted density (+/- 95% confidence interval) across the range of percent boulder, holding other factors in the model constant.

\(~\)

\(~\)

4.5.8 Short Red Gorgonian (Swiftia spauldingi)

The Short Red Gorgonian is a focal species for only the ROV surveys because its distribution corresponds better with the ROV’s deeper sampling depths. They were subsampled from video frames using an intercept method different from that used for other invertebrates (see Methods above).

Substantial reduction in densities of Short Red Gorgonians across the sampling period.

Short Red Gorgonian densities were substantially lower in 2017 than in 2012 across sites (p < 0.05, Fig. 18, Table 30). There was no significant difference in density among sites, though the Cascade Head Marine Reserve tended to have some of the highest densities in 2012. The selected GAM model included a single smooth effect of depth and a single linear effect of percent boulder across Sites. The selected model was:

Count = Year + Site + s(Depth, k = 3) + Boulder.pct, offset = log(area), family = nb

GAM model results can be found in the links below:

Short Red Gorgonian density had a unimodal relationship with depth across sites, peaking in density at around 35 m depth, though the estimated smooth had a very broad confidence interval at greater depths (p < 0.05, Fig. 18, Table 32). The relationship with percent boulder was nonsignificant.

4.5.8.1 Short Red Gorgonian Density Timeseries

Fig. 18: Short Red Gorgonian density with 95% confidence intervals at the Cascade Head Marine Reserve and its associated comparison areas. Data are See separate tabs for density timeseries and plots of GAM covariate smooths.

Fig. 18: Short Red Gorgonian density with 95% confidence intervals at the Cascade Head Marine Reserve and its associated comparison areas. Data are See separate tabs for density timeseries and plots of GAM covariate smooths.

4.5.8.2 GAM smooth for Depth

Fig. 18: Short Red Gorgonian GAM smooth for depth at the Cascade Head Marine Reserve and its associated comparison areas. The value is the predicted density (+/- 95% confidence interval) across the range of depths, holding other factors in the model constant.

Fig. 18: Short Red Gorgonian GAM smooth for depth at the Cascade Head Marine Reserve and its associated comparison areas. The value is the predicted density (+/- 95% confidence interval) across the range of depths, holding other factors in the model constant.

\(~\)

\(~\)

4.6 Additional Species Density

4.6.1 Burrowing Sea Cucumber (Cucumaria sp.; Eupentacta sp.)

Highly variable densities of Burrowing Sea Cucumbers among transects, with different trends at different sites.

Burrowing Sea Cucumber densities were characterized by high variability among transects, sites, and years, contributing to a complicated Site * Year interaction (Fig. 19, Table 33), making broad interpretation difficult without further analysis within individual sites and years. In general, Burrowing Sea Cucumber density increased at the Cascade Head Marine Reserve across the study period, while the opposite trend was observed at the Cavalier Comparison Area. The selected GAM model included individual smooth effects of depth and percent boulder at each site. The selected model was:

Count = Year + Site + Year * Site + s(Depth, by = Site, k = 3) + s(Boulder.pct, by = Site, k = 3), offset = log(area), family = nb

GAM model results can be found in the links below:

At the Cascade Head Marine Reserve and the Cavalier Comparison Area, Burrowing Sea Cucumber density increased with depth past approximately 30 - 35 m, with slightly different nonlinear responses, while at the Schooner Creek Comparison Area Burrowing Sea Cucumber density peaked at around 30 m depth (Fig. 19, Table 35). Burrowing Sea Cucumber density had distinct nonlinear responses to percent boulder at the Cavalier Comparison Area and the Schooner Creek Comparison Area (Fig. 19, Table 35), while the response at the Cascade Head Marine Reserve was nonsignificant.

4.6.1.1 Burrowing Sea Cucumber Density Timeseries

Fig. 19: Burrowing Sea Cucumber density with 95% confidence intervals at the Cascade Head Marine Reserve and its associated comparison areas. Data are See separate tabs for density timeseries and plots of GAM covariate smooths.

Fig. 19: Burrowing Sea Cucumber density with 95% confidence intervals at the Cascade Head Marine Reserve and its associated comparison areas. Data are See separate tabs for density timeseries and plots of GAM covariate smooths.

4.6.1.2 GAM smooth for Depth

Fig. 19: Burrowing Sea Cucumber GAM smooth for depth at the Cascade Head Marine Reserve and its associated comparison areas. The value is the predicted density (+/- 95% confidence interval) across the range of depths, holding other factors in the model constant.

Fig. 19: Burrowing Sea Cucumber GAM smooth for depth at the Cascade Head Marine Reserve and its associated comparison areas. The value is the predicted density (+/- 95% confidence interval) across the range of depths, holding other factors in the model constant.

4.6.1.3 GAM smooth for boulder

Fig. 19: Burrowing Sea Cucumber GAM smooth for percent boulder at the Cascade Head Marine Reserve and its associated comparison areas. The value is the predicted density (+/- 95% confidence interval) across the range of percent boulder, holding other factors in the model constant. The display of confidence intervals (gray bands) has been truncated by limiting the y axis in order to aid visualization of the smooth curves; the missing confidence intervals would extend far above the top of the displayed plot.

Fig. 19: Burrowing Sea Cucumber GAM smooth for percent boulder at the Cascade Head Marine Reserve and its associated comparison areas. The value is the predicted density (+/- 95% confidence interval) across the range of percent boulder, holding other factors in the model constant. The display of confidence intervals (gray bands) has been truncated by limiting the y axis in order to aid visualization of the smooth curves; the missing confidence intervals would extend far above the top of the displayed plot.

\(~\)

\(~\)

4.6.2 Fish Eating Anemone (Urticina piscivora)

Increasing density of Fish Eating Anemones across the sampling period.

There was no overall difference in Fish Eating Anemone density among sites (Fig. 20, Table 36), though the highest mean density was observed at the Schooner Creek Comparison Area.

At the Cascade Head Marine Reserve, density was higher in 2013 and 2017 than in 2012 (p < 0.05, Fig 20, Table 36, Table 37). The selected GAM model included a Year * Site interaction (although it was non-significant), a single smooth effect of depth across sites, and a single smooth effect of percent boulder across sites. The selected model was:

Count = Year + Site + Year * Site + s(Depth, k = 3) + s(Boulder.pct, k = 3), offset = log(area), family = nb

GAM model results can be found in the links below:

Across sites, a trend toward increasing Fish Eating Anemone density with increasing depth was nonsignificant (Fig. 20, Table 38), but density increased relatively linearly with increasing percent boulder (p < 0.05, Fig. 20, Table 38).

4.6.2.1 Fish Eating Anemone Density Timeseries

Fig. 20: Fish Eating Anemone density with 95% confidence intervals at the Cascade Head Marine Reserve and its associated comparison areas. Data are See separate tabs for density timeseries and plots of GAM covariate smooths.

Fig. 20: Fish Eating Anemone density with 95% confidence intervals at the Cascade Head Marine Reserve and its associated comparison areas. Data are See separate tabs for density timeseries and plots of GAM covariate smooths.

4.6.2.2 GAM smooth for Depth

Fig. 20: Fish Eating Anemone GAM smooth for depth at the Cascade Head Marine Reserve and its associated comparison areas. The value is the predicted density (+/- 95% confidence interval) across the range of depths, holding other factors in the model constant.

Fig. 20: Fish Eating Anemone GAM smooth for depth at the Cascade Head Marine Reserve and its associated comparison areas. The value is the predicted density (+/- 95% confidence interval) across the range of depths, holding other factors in the model constant.

4.6.2.3 GAM smooth for boulder

Fig. 20: Fish Eating Anemone GAM smooth for percent boulder at the Cascade Head Marine Reserve and its associated comparison areas. The value is the predicted density (+/- 95% confidence interval) across the range of percent boulder, holding other factors in the model constant.

Fig. 20: Fish Eating Anemone GAM smooth for percent boulder at the Cascade Head Marine Reserve and its associated comparison areas. The value is the predicted density (+/- 95% confidence interval) across the range of percent boulder, holding other factors in the model constant.

\(~\)

\(~\)

4.6.3 Basket Star (Gorgonocephalus eucnemis)

Higher density of Basket Stars at the Cascade Head Marine Reserve than at the two comparison areas.

Basket Star density was higher at the Cascade Head Marine Reserve than at either of the two comparison areas (p < 0.05, Fig. 21, Table 39, Table 40).

Increasing density of Basket Stars across the sampling period.

At the Cascade Head Marine Reserve, Basket Star density was higher in 2017 than in 2012 (p < 0.05, Fig. 21, Table 39, Table 40). The selected GAM model included individual smooth effects of depth at each site, and a single linear effect of percent boulder across sites. The selected model was:

Count = Year + Site + s(Depth, by = Site, k = 3) + Boulder.pct, offset = log(area), family = nb

GAM model results can be found in the links below:

Basket Stars had distinct relationships with depth at the three sites. At the Cascade Head Marine Reserve and the Cavalier Comparison Area, densities peaked near 35 m depth and declined in deeper waters, while at the Schooner Creek Comparison Area density plateaued past 35 m, although with substantial variability in deeper areas.(p < 0.05, Fig. 21, Table 41).

4.6.3.1 Basket Star Density Timeseries

Fig. 21: Basket Star density with 95% confidence intervals at the Cascade Head Marine Reserve and its associated comparison areas. Data are See separate tabs for density timeseries and plots of GAM covariate smooths.

Fig. 21: Basket Star density with 95% confidence intervals at the Cascade Head Marine Reserve and its associated comparison areas. Data are See separate tabs for density timeseries and plots of GAM covariate smooths.

4.6.3.2 GAM smooth for Depth

Fig. 21: Basket Star GAM smooth for depth at the Cascade Head Marine Reserve and its associated comparison areas. The value is the predicted density (+/- 95% confidence interval) across the range of depths, holding other factors in the model constant.

Fig. 21: Basket Star GAM smooth for depth at the Cascade Head Marine Reserve and its associated comparison areas. The value is the predicted density (+/- 95% confidence interval) across the range of depths, holding other factors in the model constant.

\(~\)

\(~\)

5 References

Anderson M.J., Walsh D.C.I. 2013. PERMANOVA, ANOSIM, and the Mantel test in the face of heterogeneous dispersions: What null hypothesis are you testing? Ecological Monographs 83(4): 557-574.

Chao A., Gotelli N.J., Hsieh T.C., Sander E.L., Ma K.H., Colwell R.K., Ellison A.M. (2014) Rarefaction and extrapolation with Hill numbers: A framework for sampling and estimation in species diversity studies. Ecol Monogr 84:45–67

Clarke K.R., Chapman M.G., Somerfield P.J., Needham H.R. (2006). Dispersion-based weighting of species counts in assemblage analyses. Mar Ecol Prog Ser 320: 11-27.

Green, R. H., & Young, R. C. (1993). Sampling to detect rare species. Ecological Applications, 3(2), 351-356.

Hill M.O. (1973) Diversity and Evenness : A Unifying Notation and Its Consequences. Ecology 54:427–432.

Hinkle D.E., Wiersma W., Jurs S.G. Applied Statistics for the Behavioral Sciences. 5th ed. Boston: Houghton Mifflin; 2003

Hsieh, T. C., Ma, K. H., & Chao, A. (2016). iNEXT: an R package for rarefaction and extrapolation of species diversity (H ill numbers). Methods in Ecology and Evolution, 7(12), 1451-1456.

Legendre P. and Anderson M.J. 1999. Distance-based redundancy analysis: Testing multispecies responses in multifactorial ecological experiments. Ecological Monographs 69(1): 1-24.

Lester, S. E., Halpern, B. S., Grorud-Colvert, K., Lubchenco, J., Ruttenberg, B. I., Gaines, S. D., … & Warner, R. R. (2009). Biological effects within no-take marine reserves: a global synthesis. Marine Ecology Progress Series, 384, 33-46.

Love, M. S., & Yoklavich, M. M. (2006). Deep rock habitats. In The ecology of marine fishes (pp. 253-266). University of California Press.

Oksanen, J. et. al. (2020). vegan: Community Ecology Package. R package version 2.5-7.

Wood, S.N. (2011) Fast stable restricted maximum likelihood and marginal likelihood estimation of semiparametric generalized linear models. Journal of the Royal Statistical Society (B) 73(1):3-36.