Madeleine Cortés
BIOE 161/L
October 6, 2011
Field Report No. 1:
Qualitative Sampling Methods
Abstract
The collection of distribution and abundance data is integral to the study of interspecies and intraspecies interactions, such as recruitment, competition, succession, and predator-prey interactions within a community. Two general sampling types are available: qualitative and quantitative methods. In this report, I evaluate our test of a qualitative SCUBA sampling method and find that it adds a large factor of buddy-pair variance to our collected data. The causes for this variance fall into two categories: diver perception and species suitability. The affects of diver perception are due to a lack of common diver training and briefing prior to data collection, while the affects of species suitability are due to including species that are inconspicuous due to their size, lack of mobility, or rarity.
Introduction
Determining the distribution and abundance of species in a community gives insight into species interactions occurring within that community and can often be extrapolated out to include a much broader range. Distribution and abundance data is essential to the detection of predator-prey interactions (Sala and Graham 2002), competition (Hallacher and Roberts 1985), recruitment (Watanabe 1984), as well as succession (Reed and Foster 1984). Understanding these interactions between species is essential for scientists and wildlife managers to preserve the natural state of wildlife communities.
There are two general types of methods for collecting subtidal distribution and abundance data using SCUBA: qualitative sampling and quantitative sampling. There are advantages and disadvantages to both types of sampling methods. The advantage to qualitative sampling is the speed and efficiency with which it can be carried out. This benefit, however, is at the cost of both precision and accuracy of the data collected. Quantitative sampling, on the other hand, provides both precise and accurate data, due to the use of detailed sampling methods such as: counts within a quadrat, random and uniform point contacts, as well as counts within a set area. The disadvantage of quantitative methods, however, is the increased underwater time, which may also necessitate other resources, such as boats to bring in additional tanks.
In this study, we utilized SCUBA and qualitative sampling methods to determine the distribution and abundance of 28 species in Monterey, California. We set out to ascertain if qualitative sampling is an appropriate sampling method to determine distribution and abundance and, more specifically, to answer the following questions. Firstly, does the data collected by a buddy pair differ between individuals? What reasons might account for these differences? Secondly, which species seem to be relatively good or poor candidates for qualitative sampling methods? Do species possess characteristics that cause them to be good or poor candidates?
Methods
We performed our qualitative surveys in the Marine Life Refuge at Hopkins Marine Station in Monterey, California. Hopkins Marine Life Refuge has a permanent transect cable, running roughly north to south, through rocky reef habitat with intermittent sandy patches. Data was collected on 28 species from three taxa: algae and plants, fishes, and invertebrates (see Fig. 1).
Firstly, to determine if data collected by each individual within a buddy pair differed, each diver (there were 28 divers in total) in each pair performed our own survey so that comparisons could be made between our data. Additionally, in order to compare the variance of the “buddy effect” to other sources of variance, we collected data at multiple locations along the permanent transect as well as at two opposing depths on either side of the transect cable. Specifically, we collected data in pairs at unique meter-marked survey areas along the transect cable. We each completed a two-leg 30-meter transect first on the deep side (90° due east) an then the shallow side (270° due west) of the permanent cable. We recorded the depth at the start and finish of each leg of each transect. We then graphically analyzed the collected data to determine the affect of the individual perception of each buddy in a pair as well as to compare that variance to the variance found at varying depths and locations.
Secondly, to test the appropriateness of qualitative sampling for each species, we each assigned a relative abundance category to each of the 28 species on our data sheets. We chose from the categories: (1) absent, (2) rare, (3) present, (4) common, or (5) abundant. We repeated this categorization on each leg of our transect in each direction from the cable. We charted and analyzed the data to determine which species appear to be relatively good or poor candidates for the use of qualitative sampling methods. We compared the relative difference in collected abundance data between buddy pairs for each species to ascertain for which species the most precise qualitative data was collected.
Results
Using variance component analysis, we compared the percentages of variance among data collected between buddies, depths, and among varying locations along the permanent transect (Fig. 2). In response to our first question, we found that the percent variance of data collected between buddies was 37%, which was higher than the variance due to depth (23%), and only slightly lower than the variance between locations along the transect (42%). Due to the small sample area in which we collected our data, we are unable to calculate the significance of the differences between the three sources of variance(no- this test is not about that. It just shows the proportion on the variance that is explained by each factor. We are not testing differences.). It seems clear, however, that the level of variance between buddy pairs in relation to the other sources of variance creates noteworthy doubt as to the accuracy of all the data collected. Due to the high level of “buddy variance”, the possibility of our repeating the same results is low. Coupled together with the low resolution of the categorization method used in qualitative sampling, the lack of repeatability indicates low precision in our data.
We additionally charted the relative difference in data between buddies against the mean abundance numbers recorded in the data[jf1] (Fig. 3). This gave insight into our second question, by showing that the relative difference in data was lowest for abundant species and peaked for species marked by surveyors as rare. We again used variance component analysis to view the variance between the three sources of variance: “buddy effect”, depth, and location along the cable (Fig. 4). We split the analyses into three graphs to look at the variance that occurred in the data collected for each taxon: algae and plants (Fig. 4a), fishes (Fig. 4b), and invertebrates (Fig. 4c). This chart gives insight into our second question by showing which taxa were most susceptible to the “buddy effect”. The percentage of variance in data between buddy pairs was highest for invertebrates, followed by algae and plants, while fishes had relatively low percent variance in buddy data. We also charted the relative difference between buddies’ data for each species grouped by taxa (Fig. 5). This gives a more specific answer to our second question by telling us which species had smaller or larger differences in data between buddies. While the fishes taxon had the lowest variance in buddy-pair data on the whole (17%, Fig. 4b), there were species within the invertebrate taxon that showed a very small relative difference in data between buddies (as low as 9%, Fig. 5c).
Discussion[jf2]
At the start of this study, we asked if buddy pairs differed between individuals in their qualitative sampling data. Our results showed, through variance component analysis (Fig. 2), that the variance percent between buddy-pair data was relatively high when compared to the factors of depth and location along the transect cable. Although variance due to location along the transect cable was slightly higher than variance due to the “buddy effect”, we would expect accurate data to show close to zero percent variance between buddy pairs and relatively high percent variance between meter-marks along the transect cable and even possibly between depths. The high percent variance in buddy-pair data shows that, by using qualitative sampling methods, we have introduced an additional variable to our data that obscures the naturally existing variables we would normally wish to explore in a study.
There are several different factors that may be responsible for this added variable, including: relative level of familiarity of each individual with the species surveyed and their usual abundance, differing interpretations of the categories used to label abundance, actual abundance of the survey species, as well as size and mobility of the survey species. The relative difference between buddy data (Fig. 3) peaked at the mean abundance category of “rare”, showing that we had the most variation in our buddy-pair data for species we classified as rare. This result indicates both the possibility that our individual perceptions of what constituted rarity in a particular species differed significantly and that our understanding of the term “rare” (versus the category for the next higher abundance “present”) may also have varied.
The second question we asked was: which species seemed to be relatively good or poor candidates for qualitative sampling methods? Species that were consistently marked as absent often had the lowest incidence of variation between buddies. For instance, the invertebrate species Haliotis rufescens (red abalone), Strongylocentrotus purpuratus (purple sea urchin), and Urticina lofotensis (white-spotted rose anemone) were all marked as absent (Fig. 6c) and had the lowest relative difference in buddy-pair data of all species (Fig. 5c). These results indicate that absence makes a species a good candidate for qualitative sampling, due to the increased accuracy of data collected on absent species.
Additionally, the results of our relative difference versus mean abundance chart (Fig. 3) indicate that rarer species were often overlooked by one of the individuals in our buddy pairs. It is possible that, overall, species that occur at low abundances (“rare” species) are poor nominees for qualitative sampling. These results, however, must be considered in tandem with the variance component analysis performed by taxon (Fig. 4), since it is clear that species in the fishes taxon had the lowest “buddy effect” and yet most species of fish had mean abundance categories of “rare” (Fig. 6).
One reason that fishes may have been exempt from buddy-pair data variance is their mobility, which is unique in comparison to all algae and most invertebrates on our data sheets. Those invertebrates that are motile move at a pace far less likely to catch our eyes as we performed the surveys. It seems fair to conclude that high mobility is a characteristic that makes species good candidates for qualitative sampling techniques. Additionally, larger size may play a role in the increased precision and accuracy of the data we collected as, overall, data on both algae and fishes exhibited lower levels of variance (Fig. 4a/b) in comparison to invertebrates, which are generally smaller in size.
It appears that qualitative sampling methods are far less reliable than I originally thought they would be, based upon the amount of variance between individual’s data within each buddy pair. This lack of reliability implies a low level of accuracy in the data we collected. While it might be useful to have qualitative data as a general indication of what species inhabited a particular community over long stretches of time, using this method to watch for specific changes in species distribution and abundance over shorter periods of time, such as through a cycle of ENSO events would not be appropriate, due to its inaccuracy.
In conclusion, we found that qualitative sampling introduces a large factor of buddy-pair variance to data collection. The causes for this variance can be divided into two categories: diver perception and species suitability. The affects of diver perception could be reduced by common diver training and briefing prior to data collection, while the affects of species suitability could be reduced by only using qualitative sampling methods for species that are conspicuous due to their large size, mobility, or their apparent absence.
Bibliography
Hallacher, L. E., and D. A. Roberts. 1985. DIFFERENTIAL UTILIZATION OF SPACE AND FOOD BY THE INSHORE ROCKFISHES (SCORPAENIDAE, SEBASTES) OF CARMEL-BAY, CALIFORNIA. Environmental Biology of Fishes 12 (2):91-110.
Reed, D. C., and M. S. Foster. 1984. THE EFFECTS OF CANOPY SHADING ON ALGAL RECRUITMENT AND GROWTH IN A GIANT-KELP FOREST. Ecology 65 (3):937-948.
Sala, E., and M. H. Graham. 2002. Community-wide distribution of predator-prey interaction strength in kelp forests. Proceedings of the National Academy of Sciences of the United States of America 99 (6):3678-3683.
Watanabe, J. M. 1984. THE INFLUENCE OF RECRUITMENT, COMPETITION, AND BENTHIC PREDATION ON SPATIAL DISTRIBUTIONS OF 3 SPECIES OF KELP FOREST GASTROPODS (TROCHIDAE, TEGULA). Ecology 65 (3):920-936.
Results (25)
___4_/4 Figure legends Accurate
___3_/4 Figure Legends well composed (complete and concise)
__5__/5 Results organized according to questions
__4__/4 Graphs presented in a logical order, case made for the order
_4___/4 Grammar, sentence structure and spelling
__3__/4 Clarity and conciseness of writing
Discussion (25)
____/9 How well did they answer the questions they present in the Intro?
1)__3__/3 Discuss the results from the specific to the general.
2)___3_/3 Do these results surprise you? In other words, is the qualitative method more or less reliable than you thought it would be, and do you think that degree of reliability (which can be assessed based on relative difference between buddies) implies anything about accuracy?
3)__3__/3 Do you think the qualitative sampling approach is appropriate for describing trends of species abundances through time? Explain your answer
__3__/3 Grammar and Spelling
__2__/2 General Thoughtfulness
__3__/3 Clarity and conciseness
__5__/5 Organization of discussion
__3__/3 Context and Bigger Picture
General Notes: The discussion should be less focused on the details and instead emphasize the main point of the study…using the details as support when needed.
- 1 -
[jf1]These statements are for the methods. In the results, state the result.
[jf2]The discussion should be less focused on the details and instead emphasize the main point of the study…using the details as support when needed.