Are qualitative studies effective in sampling for abundance in a kelp forest?
Title: 4 pts
Lauren Smith
Bio 161 Kelp Forest Ecology
University of California, Santa Cruz
Abstract [2, 0, 2, 2, 2, 2] 10 pts
Species distribution and abundance data can be collected using qualitative or quantitative methods. This study tests the validity of a qualitative study performed using SCUBA in a kelp forest. Buddy teams gathered data on species abundances along a 30-meter transect and this these data was were analyzed for discrepancies. The data differed greatly between buddy pairs working on the same study site therefore this method is not the most reliable for sampling abundances. [no… it was the fact that estimates differed as much or more between individuals in a pair as they did among pairs across the reef that raises concern.]
Introduction [2, 4, 4, 2, 2, 2, 4, 0, 2,2, 2] 26 pts
Distribution and abundance studies can lead to inferences made about the coexistence of species in a community, patterns, links, and structures within ecosystems, or competition and predator relationships. In kelp forests, competition for light, changes in depth, predation, and competition affect the distribution and abundance of species (Reed and Foster 1984 and Watanabe 1984). Surveying this information (??) can be used to show differences and predict changes after disturbances, climates changes, and regime shifts (Ebeling et. al. 1985, Anderson and Piatt 1999, and Warren et. al. 2001). There are many techniques to sample for distributions and abundance of species but most fall under two categories: qualitative or quantitative. This was a qualitative experiment [no… we did not do an experiment… study] done using SCUBA comparing buddy pairs assessment of species abundance along a transect in effort to discover if qualitative methods are reliable when collecting distribution and abundance data.
Qualitative and quantitative sampling techniques each have strengths and weaknesses. In an underwater environment time is valuable and studies are limited to air consumption and nitrogen on gassing. (??) Qualitative surveys are well designedmight be appropriate for such a study because they are generally quicker than quantitative surveys. Qualitative studies, however, are also more subjective than quantitative surveys. Quantitative studies are objective, easier to generalize, and all aspects are designed before the experiment study commences. The validity of a qualitative survey depends on the skill and thoroughness of the researcher where quantitative studies depend more on the measurement device and instruments used.
This qualitative study, conducted in the field, was designed to illustrate determine the differences in the sampling abundances estimates between pairs working at the same location and discover identify the species better suited for qualitative sampling methods. This leads to inferences made about characteristics or traits that may make a species a better or worse candidate for a qualitative study. [so what were the various questions the study was designed to answer??]
Methods [2, 2, 0, 0, 2, 2, 2, 0, 2, 0, 2, 2, 4, 2, 4, 2, 2] 30 pts
Insert a statement about the overall approach here.
Study Sitesystem
Qualitative observations were made off the Hopkins Marine Station located in Pacific Grove, California (36˚37’16.12’’N, 121˚54’13.88’’ W) in September, 2011. Offshore there is a permanent cable running about parallel to the beach from east to west. Every 10 meters is marked and has an eyebolt for running transects perpendicular to the cable (on or offshore). [Is it a kelp forest? How deep? What kind of rock? Temperature?]
Collection of Data
Fourteen buddy pairs equipped with SCUBA were randomly assigned meter marks off the permanent cable in 5 meter increments (those not on the 10 meter marks measured out 5 meters to their assigned number) and attached their transect lines to the cable. At each site four sets of data were taken; the first while running the meter tape out on a 90˚ heading offshore (leg 1), the second while reeling the meter tape back towards the cable (leg 2), and the last two used the same methods on the 270˚ onshore site (legs 3 and 4 respectively). For each leg each of the 28 target species found within two meters of the tape (1 meter to either side of the tape per buddy) was given a number code 1 through 5 corresponding with its distribution at that site; 1= absent, 2= rare, 3= present, 4= common, and 5= abundant. The target species included five species of algaes—Cystoseria osmundacea, Chondrocanthus corymbiferus, Dictyoneurum californica, Macrocystis pyrifera, Dictyoneuropsis reticulata—one angiosperm—Phyllospadix spp.—nine species of fish—Oxylebius pictus, Hexagrammos decagrammus, Sebastes mystinus, S. carnatus, S. chrysomelas, S. atrovirens, Embiotoca jacksoni, E. lateralis, Damalichthys vacca—and thirteen species of invertebrates—Patiria miniata, Pycnopodia helianthoides, Pisaster brevispinus, Pisaster giganteus, Urticina picivora, U. lofotensis, Pachycerianthus fimbriatus, Balanophyllia elegans, Tethya aurantia, Calliostoma ligatum, Haliotis rufenscens, Strogylocentrotus fransiscanus.
[better to use questions as subheadings and combine design, methods, predictions and analyses for each question.]
Analysis
In order to estimate the difference between buddy pairs we calculated the relative difference (in percentages) between buddy pairs as a function of the mean abundance of the species (codes 1 through 5).
Next we calculated the percent of the buddy pairs that disagreed on the presence or absence of a species. These variances were compared on a species by species level providing a basis to which species are better candidates for qualitative studies.
We also looked at the relative difference between buddies (in percentages) for each species based on code data. This allows us to detect species the researchers were able to identify well and poorly during the study, which had the lowest versus highest differences. In order to see if the presence or absence of each species played a role in the discrepancies within buddy pairs we averaged the mean abundance for each species compared species with low and high disagreements in abundances to their mean abundances.
Results
The difference between buddy pairs when comparing mean abundance with the relative differences between buddies (in percentages) is greatest between codes 1, 2, and 3 (Figure 1).
There are only two species that had no variance between present and absent; all buddy teams agreed that giant kelp, Macrocystis pyrifera, and bat stars, Patriria minitata were present. The fish with the lowest variance was the black and yellow rockfish (S. chrysomelas) and the highest was the striped surfperch (E. lateralis). D. reticulata and D. californicum had the highest and second highest disagreements, respectively, of the algae species. The sand anemone (P. fimbriatus) had the highest disagreement with in the inverts (Figure 2).
Fish and algae species with the highest and lowest disagreement between buddies (in percent) based on code data (1-5) is consistent with the variance found in the present/absent disagreement calculations (Figure 3). Comparing disagreements within invertebrates, however, red abalone (H. rufescens), white-spotted rose anemone (U. lofotensis), red urchin (S. fransiscanus), and fish eating anemone (U. picivora) each have lower percent disagreement than bat stars. The invertebrate with the highest difference was the orange cup coral (B. elegans).
Fish had the lowest mean abundances; each had a mean abundance below two (rare). Algae and invertebrates, however, ranged from absent to abundant (Figure 4).
Discussion
Buddy pair sampling data differed during the survey. The biggest differences were found between codes 1, 2, and 3 (absent, rare, and present). This could be caused by a number of factors; using different definitions for rare and present, lack of knowledge of the target species or the presence of only one organism making it rare for one buddy and absent for the other.
If only concerned with the presence or absence of a species the best candidates would be giant kelp and bat stars (Figure 2). This is because they were common and easy to identify. Looking at abundances, however, the best candidates, or species with the lowest discrepancies between buddies are, red abalone, red urchin, fish eating anemone, and black and yellow rockfish (Figure 3). Giant kelp and bat stars had no discrepancies between buddies marking them present or absent, but when comparing species abundance variance using the codes, these two species no longer have the smallest difference. This shows the researchers’ definitions for the codes range. The red abalone, red urchin, fish eating anemone and black and yellow rockfish make up some of the lowest mean abundance values, this suggests that the variance is small for these species because they were not present. I do not believe that these species had any distinguishing features making them easier to identify. They only appear to be the best candidates because they were often absent from the study site.
The poorest candidates, species with the highest discrepancies (Figure 3), are D. reticulata, D. californicum, stripped surfperch, kelp greenling (H. decagrammus), painted greenling (O. pictus), orange cup coral, ring topped snail (C. ligatum), sand anemone, great spined star (P. giganteus) and orange puff ball sponge (T. aurantia). The high amount of variance in D. reticulata and D. californicum may be partially explained by the data sheet listing them both under the genus Dictyoneurum causing the researchers to become confused. The high variance in the fish species is likely due to their ability to move quickly and the difficulty of taking abundance data in the benthos as well as in the water column while swimming along a transect. Orange cup coral, ring topped snail, sand anemone, great spined star and orange puffball sponge all have abundances over 2 (rare), meaning, that like giant kelp and bat stars, the variance between buddies for these species may be due to differences in code definitions between individuals in a pair.
This method of qualitative sampling does not seem like the best approach for collecting data on species abundance. This is not surprising in this instance because the definitions for the codes (absent to abundant) were not discussed prior to the start of the study and not every researcher was able to positively identify each of the listed target species.
I do not believe any of the information found in this study to be reliable due to the subjective nature of the study, except if we had looked only at the presence versus absence of each species. This would not provide abundances, but may give an idea of which target species are found at Hopkins during the fall.
This qualitative sampling approach may be effective for describing trends of species over time if the same researchers are used consistently and therefore the codes will keep the same definitions with each new survey. If the codes were more specific and the researchers better capable of identifying the species, this study may have been improved.
References [3, 3] = 6 pts
Anderson, P.J and J.F. Piatt. 1999. Community reorganization in the Gulf of Mexico following ocean climate regime shift. Marine Ecological Progress Series 189:117-123.
Ebeling, A.W., D.R. Laur and R.J. Rowley. 1985. Severe storm disturbances and reversal of community structure in a southern California kelp forest. Marine Biology 84:287-294.
Reed, D. C. and M. S. Foster. 1984. The effects of canopy shading on algal recruitment and growth in a giant kelp forest. The Ecological Society of America 64(3):937-948.
Warren, M. S., J.K. Hill, T.J. Asher, R. Fox, B. Huntly, D.B. Roy, M.G. Telfer, S. Jeffcoate, P. Harding, G. Jeffcoate, S.G. Willis, J.N. Greatorex-Davies, D. Moss and C.D. Thomas. 2001. Rapid responses of British butterflies to opposing forces of climate and habitat change. Nature 414:65-69.
Watanabe, J. M. 1984. The influence of recruitment, competition, and benthic predation on special distributions of three species of kelp forest gastropods (trochidae: Tegula). Ecology 65 (3):920-936.