Kelp Forest Ecology

Ian Jacobson

Kelp Forest Ecology

29th October 2011

Total= 4+20+34+37

Lit Cited= 5

[4]Comparison of two protected kelp forests in central California and their associated sources of variability in community composition

Abstract [20]

First one sentence on why this was an important study…With a quantitative sampling approach, two kelp forest communities in Monterey Bay were compared to investigate differences in community composition. Differences in community composition and species assemblages from three taxonomic groups, fish, algae, and invertebrates, were associated with[this is a result… first state the hypotheses or questions you were testing] variability explained by site, sampling day, and an interaction effect from both factors. Species abundance data was gathered on SCUBA and compared to identify primary sources of variability. We found that the community composition between Hopkins and Pt. Lobos was significantly different and this was attributed to variability associated between sites. Algae and invertebrate species assemblages were also significantly different between sites indicating site variability drove this difference. Fish species assemblages was found to have a significant sampling day effect and an interaction effect meaning that assemblages differed, but sampling day overshadowed the differences seen between sites. We concluded that differences seen between sites was due to difference in habitat and exposure, while difference between sampling days was attributed to the different conditions, with respect to swell height.

Introduction [3, 4, 3, 3, 3, 0, 4, 3, 4, 3, 3]= 34

Similar ecosystems that exist within a region can have very different community compositions. Dynamic interactions drive thesedifferencesat a broad range of spatial and temporal scales (Reed et al, 1988;Edwards, 2004). Understanding habitat heterogeneity is useful when looking at patterns across habitats as their composition and configuration affects species abundance over time and space (Connell and Irving, 2008). For example, aA study comparing areas inside and out of a protected area by Ling et al (2009) uncovered the importance of protecting apex predators due to maintenance effect on kelp forest habitats. By studying differences across habitats important ecological processes can be foundidentified.

Along the west coast North America, kelp forest ecosystems are highly diverse and productive, especially along the California coastline (McClanahan and Branch, 2008). These ecosystems provide ecologists with the opportunity to study the diverse communities associated with the habitat forming algae giant kelp, Macrocystis pyrifera. With a quantitative approach, ecologists are able to estimate populations to describe community composition across multiple sites, but must be careful to accurately interpret findings due to the complexity of these systems. Differences found between sites can be explained by spatial and temporal variability, whether they interact or affect the system separately, between community compositions.

In this study we are comparing two protected kelp forest communities in central California to look for differences in community composition and the associated causes. The purpose of this study is toidentify if the community compositions of Point Lobos and Hopkins differ. [[better to pose all of these predictions as questions… if you make such predictions, you need to explain why you are predict those particular outcomes!] We expect predict species composition between the two sites to be different. [why????] We also investigate if species composition varies between three major taxonomic groups within each site. Differences in species abundance are expected predicted within each taxonomic group between each site. [why????] To identify if there is a temporal effect in our sampling, we investigate if the species composition of each site is different between two sampling days. We do not expect predict to find a difference in composition within each site between sampling days due to the short time interval between sampling. We further investigate ifdifferences in species abundance vary between taxonomic groupsfor each site between sampling days. Differences within each taxonomic for each site are not expected predicted to differ between sampling days. Finally, we will determine if there is an interaction effect of site and sampling day that affects species composition. No interaction affect is expected predicted in species composition. We further look test forto see if there is an interaction affect that varies between taxa. No interaction affect across taxa is expectedpredcited.

Methods [2, 2, 4, 3, 3, 2, 1, 2, 2, 0, 3, 0, 3, 2, 3, 2, 3] 37

To detect spatial and temporal differences between two kelp forest communitieswe quantitatively sampling sampled on SCUBA collecting species abundance from select species of three major taxonomic groups, fish, invertebrates, and algaeon SCUBA (Table 1. All species chosen were common to Monterey and easily countable.Our selection consisted of mobile and sessile species due to the likelihood possibility that we would see day-to-day variation in mobile species and little day-to-day variation in sessile species.We compared average species abundance between sites and across taxa between two kelp forest communities, sites, and between sampling days to investigate if variation can be explained temporally or spatially. [nice!]

Study System [this is where the description of what and why the study spp were selected belongs instead of above!]

Sub-tidal species abundance data was taken at protected two sites, Hopkins Marine Reserve and Pt. Lobos state reserve.[lat and lon for each site????] As described by Wantanabe (1984), Hopkins is a protected reef with shallow large granitic benches and high relief outcrops. Pt. Lobos is a much more exposed spot south of Monterey. Sampling took place two consecutive days, Tuesday October 11th and Thursday October 13th. Two groups of eight buddy pairs sampled each site on a different day to obtain data for both sites on both days. At Lobos, buddy pairs ran a 30m x 2m transect line at two meter marks separated by 10 m offshore, perpendicular to the 100m site transect. At Hopkins, each buddy pair sampled one meter mark in an offshore, 90, and Onshore, 270, direction perpendicular to the permanent site transect. Altogether, there were 16 replicate transects at each site on each day to yield mean species abundance.Mean species abundance was used to compare community composition and species assemblages between sites and across sampling days.

Differences in communitycomposition between sites

We comparedcommunitycomposition for each sitewere usinga PERMANOVA with a threshold of 0.05 was used to determine if differences were significant. [so… re word the previous sentence as such… We used the multivariate analysis PERMANOVA to test for differences in the community structure between the two study sites.]]A p-value less than 0.05 for the site factor indicates a significant difference between sites that would support our hypothesis. You do NOT need to talk about a p-value of 0.05… all ecologists know this. Disn’t you do another analyses to compare community structure between sites (e.g., MDS plots, etc.)??

Differences in species composition between sites across taxa

For each taxonomic group, mean species abundance data was compared across sites. We used a PERMANOVA with a threshold of 0.05 to determine if there were significant differences between sites for fish, algae, and invertebrates. And you compared the results of the analyses among the groups to determine A p-value less than 0.05 in the site factor indicates a significant difference.

Difference in community composition across sites between sampling days

The same PERMANOVA as above was used to determine if the community composition between the two sites varied between sampling days. A significant difference between sampling days would yield a p-value less than 0.05 in the sampling day factor. [same comments as above]

Difference in species composition across taxa between sampling days

The same PERMANOVA as above was used for each taxonomic group to determine if mean abundance differed between sampling days. For fish, algae, and invertebrates, a p-value less than 0.05 in the sampling day factor indicates a significant difference between sampling days.

Site and sampling day interaction effect on community composition

To determine if there was an interaction effect between site and sampling day on community composition, we used the same PERMANOVA as above. A significant interaction effect would yield a p-value less that than 0.05 in the site and sampling day interaction factor.

Site and sampling day interaction effect on species composition across taxa

For each taxonomic group we tested for an interaction effect between site and sampling day by using the same PERMANOVA as above. A significant interaction affect would be indicated by a p-value of less that 0.05 in the site and sampling day factor.

Results

Community composition between sites

We found a significant site effect on the community composition between Hopkins and Lobos (Fig. 1, PERMANOVA: Site; p=0.001).

Species composition between sites across taxa

We found a significant site effect on the species composition of algae between Hopkins and Lobos (Fig. 2, PERMANOVA: Si; p=0.001). For fish, we did not find a significant site effect on species composition (Fig. 3, PERMANOVA: Si; p=0.319). We also found a significant site effect on the species composition of invertebrates between Hopkins and Lobos (Fig. 4, PERMANOVA: Si; p=0.001).

Community composition between days

We did not find a significant day effect on the community composition of Hopkins and Lobos between sampling days (Fig. 1, PERMANOVA: Sampling_Day, p=0.724).

Species composition between sampling days across taxa

We found that sampling day did not have a significant effect on the species composition of algae between Hopkins and Lobos (Fig. 2, PERMANOVA: Sa;p=0.724). For the species composition of fish, we found a significant day affect (Fig. 3, PERMANOVA: Sa; p=0.043). For invertebrate assemblage, we found there to be no significant day affect on species composition (Fig. 4, PERMANOV: Sa; p=0.505).

Interaction affect on community composition

We did not find any significant interaction affect on the community composition between Hopkins and Lobos (Fig. 1, PERMANOVA: SitexSampling_Day; p=0.375).

Interaction effect on species composition across taxa

There was no significant interaction effect on the species composition of the algal assemblage (Fig. 2, PERMANOVA: SixSa; p=0.926). We sound a strong interaction effect on the species composition of fish assemblage (Fig. 3, PERMANOVA: SixSa; p=0.015). We also found no significant interaction effect on the species composition of the invertebrate assemblage (Fig. 3, PERMANOVA: SixSa; p=0.569).

Discussion

Our findings indicate that the difference in community composition between Hopkins and Lobos are driven by the site variation. We are able to fail to reject our hypothesis that community composition varies between sites and reject our hypothesis that community composition varies between sampling days indicating there is no interaction effect. This was expected due to the different habitat features both sites exhibit. Hopkins is a much more protected site while Lobos is more exposed to northwest swell lending to an increased likelihood to hydrological disturbance which influences community structure (Connell, 1978).

When examining the differences in species assemblages across taxa between Hopkins and Lobos, our findings suggest that different taxonomic groups are subject to different sources of variability.We found that the sampling site influences differences in algae and invertebrate assemblages, but differences in fish assemblages are influenced more by sampling day (Figure 5). This is to be expected because fish are mobile. Another reason for this difference can be attributed to the different conditions between sampling days. On Tuesday, October 11th, NOAA buoy readings measured swell height ranged from 1.6m-2.4m from a WNW direction while Thursday October 13th swell height ranged from 2.9m- 4.3m from the NW (NOAA, 2011). During days with significant swell, sampling for these mobile species can be difficult because they tend to seek refuge from the intense hydrodynamic activity, resulting in lower abundance counts. The interaction effect between site and day for fish means that there was a small site difference, but the variation between sampling days overwhelms this.

To improve this study a larger sample size for fish and should be taken. An analysis of power indicated that we would need more that 30 transects to yield more powerful results.More power would make our results more statistically robust. For algae, power analysis indicated that our sample size was adequate for this study (Figure 6).

Individual species yielded different power indicesfrom each taxonomic group. For fish the species that yielded the least amount of power was Sebastes chrysomelas. For Algae, Dictyoneuropsisreticulata yielded a power index of less than two. For invertebrates, Strongylocentrotusfransiscanus had the lowest power index. These species therefore contributed to most of the variability within our results indicating that these species are probably not the best species to include in our sampling. This could be attributed to the difficulty in identifying and finding these species. S. fransiscanusand S. chrysomelas are often found hidden in cracks, while D. reticulata is very similar to Dictyoneurumcalifornicum. This makes these species particularly difficult to sample.

To control for the variability we saw between sampling days, it would be better to plan dives around swell models. If we sampled on days with similar conditions, we would expect to see a shift in the variability between sampling days to variability between sites. Since there was an interaction effect, that there were differences in fish assemblages between sites, but swell had an obvious affect on our results displacing more variability to our sampling day factor.

With this quantitative approach, we were able to investigate the differences between two similar communities. We found that temporal and spatial variability influence composition of communities and their constituent taxonomic assemblages. Though for some specific taxa both temporal and spatial factors can have an interaction effect.

Figures

Table 1 List of species used, organized by taxa, to gather quantitative data on community composition and species assemblages.

Figure 1 Multi dimensional analysis (MDS) plot of community composition between Hopkins and Lobos. Each point represents one transect labeled by site and sampling day. Distance between points represents dissimilarity.

Figure 2 MDS plot of species assemblage of algae. Each point represents one transect labels by site and sampling day. Distance between points represents dissimilarity. Separate groupings of similarly labels points represents differences in algal assemblages between sites.

Figure 3 MDS plot of fish species assemblages. Each point represents one transect labeled by site and sampling day. Distance between points represents dissimilarity.

Figure 4 MDS plot of invertebrate assemblages. Each point represents one transect labeled by site and sampling day. Distance between points represents dissimilarity.

Figure 5 Associated sources of variability from algae, fish, and invertebrate species assemblages. Graphs depict sources of variability. Algae and invertebrate assemblages are influenced by site variability, while fish assemblage is affected mostly by sampling day.

Figure 6 Analysis of power for algae, fish, and invertebrate assemblages. Asymptote above a power index of 2 represents significant power. For algae species composition 25 transects is sufficient. Fish and invertebrates require more than 30 transects for sufficient power.

Literature Cited

Connell, J. H. 1978. Diversity in Tropical Rain Forests and Coral Reefs. Science 199:1302-1310

Connell, Sean S. D. andAndrew A. D. Irving. 2008. Integrating ecology with biogeography using landscape characteristics: a case study of subtidal habitat across continental Australia. Journal of Biogeography. 35:1608-1621

Edwards, Matthew M. S. 2004. Estimating scale-dependency in disturbance impacts: El Ninos and giant kelp forests in the northeast Pacific. Oceologia. 138:436-447

Ling, S. D., C. R. Johnson, S. D. Frusher, and K. R. Ridgway. 2009. Overfishing reduces resilience of kelp beds to climate-driven catastrophic phase shift. PNAS. 106:22341-22345

McClanahan, Timothy T. R. and George G. M. Branch. 2008. Food Webs and the Dynamics of Marine Reefs. Oxford University Press. New York.

NOAA. Accessed October 28th, 2011.

Reed, Daniel C., David R. Laur, and Alfred W. Ebeling. 1988. Variation in algal dispersal and recruitment the importance of episodic events. Ecological Monographs. 54:321-335

Wantanabe, James M. 1984. The Influences of Recruitment, Competition, and Benthic Predation onSpatial Distributions of Three Species of Kelp Forest Gastropods (Trochidae: Tegula). Ecology. 65:920-936