Species abundance variability between two kelp forest beds along the Central California coast: Hopkins vs Point Lobos.

Jacqueline Selbitschka

Abstract

There is a high diversity of species within the kelp forests in central California. However, not all of these kelp forests are dominated by the same species. Knowing information on species and their abundances is useful when creating MPAs that will be environmentally impactful with minimal affects on economics. This information is also useful when determining the affects climate change and anthropogenic perturbation is having on the environment. We compared the abundance of29 species spanning 3 taxonomic groups within two kelp forests in the Monterey Bay, Hopkins and Point Lobos. Both are temperate kelp forests that are dominated by the algae Macrocystis pyriferaand part of the marine protected areas (MPAs) that span the coast of California. We found that there was a site effect for all species, but different taxa acted differently and caused day effects and interaction effects. A PERMANOVA showed that fishes had a day and an interaction effect. This was probably due to its low power index and an artifact of not having enough data. The differences in these two sites are being driven by exposure to wave action due to its effects on Macrocystis pyriferasurvivorship during high energy episodes.

Introduction

Species interactions are the basis to understanding the ecology of a system. How organisms interact within a system is dependent on what species are in the system and their relative abundances. When considering conservation efforts this knowledge is needed whendetermining fishery limits and creating affective protected areas without over limiting resources that provide economic services.Along the coast of California there are many different types of kelp forest communities. Surveys of these communities and knowledge of their life history traits helped to determine the distance between MPAs to create a network that gives the most protection to populations. Marine Protected Areas (MPAs) were established based on species composition of kelp forests along the coast of California. There are different levels of protection that MPAs provide, by have the these general goals; to increase the number of large individuals and the overall size of fishery species populations, preserve relative abundances and their roles within a community and protect functional processes of an ecosystem (Carr et al., 2011). There have been areas that were once fished turned to no-take zones. Cabo Pulmo National Park (CPNP) is part of long term monitoring program had a 30% annual increase in fish biomass (Aburto-Oropeza, 2011). These efforts provide practical services such as maintaining populations of important fishery species, recreational uses and processes that have a global affect on climate.

Abundance surveys also help to understand how perturbations to oceanographic processes affect species and their abundances as well as to observe changes over time due to these factors. Climate is driven by ocean chemistry, temperature and currents. Changes to these things effect the climate and in turn the organisms that live within ocean environments. With studies of species abundances we can observe how changes to these factors are affecting ecosystems and how it could affect them with continued perturbations or implemented protection. A study done by Schiel et al. used the abundances of species to track changes in distributions due to temperature increases. This system simulated environment temperature increases and tracked changes in community structure over 18 years. This study demonstrated that it is hard to determine the way that a community with react to a change in conditions, but there was significant change in community structure that happened due to warming. This study looks at a similar system to the two sites that we compared.

We compared kelp forest communities at Hopkins and Point Lobos. These temperate kelp forests are approximately 20 miles apart along the coast of Central California. Kelp forests in this area are primarily composed of the algaeMacrocystis pyrifera, a habitat structure and food source. A vast number of invertebrates live on these plants as well as beneath the canopy and within their holdfasts (Hughes, 2010). Some fishes use it as a nursery habitat that provides food and protection (Craig, 2004). This system is diverse yet not all kelp forests have the same species composition. We wanted to determine if the kelp forests in two different locations, yet within close proximity to each other, are comprised of the same dominate species and how their abundances vary.

We tested the hypothesis that there is a difference in species abundances between Hopkins and Point Lobos. There are three types of variability that could cause a positive result for our hypothesis; spatial variability, temporal variability and unexplained variability. To determine the source of variance we break down the hypothesis into three questions: 1) Is there a difference in species composition between Hopkins and Point Lobos, 2) Is there a difference in species composition between days and 3) Do both site and sampling day affect species composition (interaction affect)?Each question is associated with a source of variability, spatial, temporal and unexplained respectively. Spatial variability will indicate that there is a real difference in species composition and abundance between sites. We want to try to limit temporal and unexplained variability. These two types improperly represent the abundances of species. Some species are more susceptible to these types of variability.

Methods

Study Sites:

The two study sites we selected are comprised of kelp forests and the species that reside within that community. The two sites are along the Central California coast, Hopkins Marine Station, Pacific Grove (36°37’N, 121°54’W) and Point Lobos, Carmel (36°30’N, 121°56’W).Hopkins was established as a marine life refuge in 1931. This limited the collection of invertebrates and algae within the designated area and later included fish. It is on the extreme end of being protected from wave action(Graham, 1997). The substratum is made up of granite outcroppingsand has more sandy patches and Diopatra beds (Jones, 1985). It has a permanent transect that runs North South. At Point Lobos we dove in the kelp forest at Whaler’s Middle Reef (fig. 1). Lobos is more susceptible to being exposed to swell than Hopkins. The substrate at Lobos is comprised of Carmelo formation and has much higher outcroppings of rocky reef (Lovejoy, 1996). Both locations had a max depth of around 40 feet at which we surveyed.



Is there a difference in species composition between Hopkins and Point Lobos?

To answer this we tested the hypothesis that there is a difference in species composition between Hopkins and Lobos. To determine the composition of species at each location we selected 29 species commonly characteristic of the Central California coast (table 1). It is easy to determine individuals of these species making them easily counted and span a large range of sizes, transportation and reproductive methods. We had 6 buddy pairs at Hopkins and 7 at Lobos on each day. At Hopkins each buddy pair was assigned a meter mark in increments of 10 to run out 30 meter transects offshore, 90°, and then onshore, 270°. Each transect is a replicate. On the way out the fish were counted within a visual box 2 meters high, 2 meters in front and 1 meter wide. The tape was then anchored and left at 30 meters. Data was collected for invertebrates and algae before reeling the meter tape back to the transect line and repeating for the onshore. At Point Lobos there is no permanent transect so a tape was laid out running East West. Each buddy pair was assigned two meter marks 10 meters apart. Both transects were run onshore, 0°, from the tape laid out. Counts for all species were broken up for every 10 meters along the 30 meter transect.

Does the difference in species composition between Hopkins and Lobos varyby taxa?

Using the data from the methods described above we split up the species by taxa and compared their relative abundances at each location to see if there was a difference in abundances between locations based on taxa.We compared the raw data between the sites and days in a four way plot. We then ran a PERMANOVA to determine if these differences were statistically significant. With this information we could also compare specific species and how well they were sampled.

Is there a difference in species composition between days?

To determine if there was a day affect species were counted at each location on two different days two days apart (October 11th and 13th). Different species have different physiological traits that make them more reactant to changes in conditions.By counting on two different days at both sites we can compare the data to see if there is a difference in relative abundances. We are not interested in a day effect, but rather if there is a day affect how that affects our data for what we are interested in, differences in community abundances between sites.

Does the difference in species composition between days vary by taxa?

To determine if there is a day affect determined by taxa we took the variance in abundances of each species by day, broke them up into their taxonomic groups and found their percent variance. We then did the same thing with the abundances of each species by site and compared them. If there is a day affect the percent variance will be higher when comparing variance in abundance between days. If there is a site effect the percent variance will be higher when comparing sites.

Do both site and sampling day affect species composition(interaction effect)?

We used the PERMANOVA for all species to determine if there was a compounded affect of both site and day or interaction effect. An interaction occurs if one of the variables differs depending on the level of another variable. We are interested in site and day variables interacting. This could cause a difference in species abundances by site that is actually confounded by day.

Does the interaction between site and sampling day vary by taxa?

To determine if there was an interaction affect for any of the taxa specifically we used the PERMANOVA that was broken down into taxa. As stated before different taxonomic groups have traits that allow for more mobility and make them able to leave an area or hide if the conditions are not favorable.

Was there enough data collected to draw accurate conclusions?

We also wanted to determine if our sampling was extensive enough to get conclusive results. We ran a power index on how many transects are needed in order to have a high enough power to have statistically significant differences. Since each taxonomic group acts differently we split them up to determine the number of transects needed for each group. The point at which the line asymptotes is the minimum number of transects needed to have powerful data. Few transects would result in inconclusive data and any more would be unnecessary.


Results[jf1]

Is there a difference in species composition between Hopkins and Point Lobos?



We found that all transects sampled at either site were more similar to other transects at that same site (fig. 3[jf2]). There was a significant difference in communities between Hopkins and Lobos (permanova all species: site effect, p=.001). What species? Contributions? Level of Difference?

Does the difference in species composition between Hopkins and Lobos varyby taxa?

We found a strong effect of site on the species composition of the algal assemblage (permanova algae: site effect, P=.001). Species contributing to this difference were Chondracanthus corybifera, Cystoseira osmundacea and Pterygophora californica. C. corybiferaand C. osmundacea has a consistently higher abundance at Hopkins and P. californica at Lobos.

Is there a difference in species composition between days?

We found that there was no day effect when considering the abundances of all species.There was some patterns of transects beingsimilar by day (fig. 3), but no statistical significance associated with their assemblages.

Does the difference in species composition between days vary by taxa?


We found that there was a significant day effect when breaking down data into taxa. Fish assemblages show a greater difference between transects on either day rather than transects between sites (permanova: sample day p=.043).For aglae and invertebrates all the variance is from the source of site and not day, where fish act different and variance come from both site and day (fig. 5).
Fish species contributing to this difference between days wereEmbiotica jacksoni, E. lateralis, Sebastes atrovirens and S. mystinus. The average number of fish per transect was very low so the differences in abundances was on a scale of 0.1 (fig. 4).


Do both site and sampling day affect species composition(interaction effect)?

We found that there was no interaction affect of day and site for all species. Overall the composition of species found at Hopkins and Lobos on day one were very similar to their respective site on day two. Give some evidence for this. Don’t just answer the question of the subheading in yes and no fashion. Give us some reasons for differences. Quantify those differences.

Does the interaction between site and sampling day vary by taxa?

We found that despite there not being an interaction affect for all species there was a significant interaction affect for fishes (permanova: sitxsampling_day, p=.015). Fishes varied both between sites and within those sites by days. The number of fishes saw on either day at either site were very low; less than one per species when averaging all the transect data.

Was there enough data collected to draw accurate conclusions?

Each taxa acted differently and therefore require a different number of transects. Algae were the only taxa that were sufficiently sampled. It was over sampled and would actually only need around 25 transects. We are unable to determine the number of transects needed to have statistically significant data for fishes and invertebrates. Invertebrates appear to have been sampled adequately around 15 transects, but then starts to increase again (fig. 6). This is a product of statistical analysis.

Discussion

We found there to be a significant difference in community composition and abundance between Hopkins and Point Lobos. To ensure that our results were significant we tested for other sources of variability that would make our results appear to be significant typically more variation leads to lower statistical power.- less ability to discern true differences. We tested for a day effect and an interaction effect. Despite fishes showing a significant day and interaction affect we determined

that our results show that there is a significant difference between sites.

Selecting both locations to be MPAs eliminated the potential compounded effects that fishing might have between the two sites since both are protected. This resulted in a stronger comparison to be made between sites. Yet, we still saw a significant day and interaction effect with fishes. As seen in this study, different taxa and abundances have different power indexes (fig. 6). These differences in power indexes are influenced by behavioral traits or mobility. Fishes had the lowest power and the lowest abundances of the taxa. Insufficient data for fishes is probably the reason for the day and interaction effect seen in the PERMANOVA.Maybe…not necessarilySome organisms such as those that are sessile require fewer replicates to get statistically significant data and less affected by changes in environmental conditions from day to day. This is opposed to active organisms which have the ability to move longer distancesand hide in crevices based on conditions affecting surge and turbidity. Preliminary studies would have allowed for us to determine how many replicates were needed in order to make our fish surveyshave a statistical power high enough, greater than 2, to make accurate conclusions.

The main difference between Hopkins and Lobos is their exposure to wave action. The intensity of exposure at a location determines the kelp cover. Macrocystis pyrifera creates drag with its abundant and large blades. These blades also reduce light availability to surrounding species. This drag, however, is detrimental to kelp survivorship when there is high wave energy (Utter and Denny, 1996). More exposed areas reduce kelp and create space and light availability for understory algal species such as Pterygophora californica and Eisenia Arborea. The stipe abundance of Macrocystis pyrifera at Hopkins was about twice that of Lobos on either day and Lobos was about 8 times more abundant in sub-canopy kelp. This could also account for the higher abundance of red algal cover at Hopkins since red algae need less light to grow.