Biostatistical Analysis Demonstrated in Cushion Sea Star Study in Graham’s Harbor

This topic submitted by Lin ( at 8:10 PM on 8/3/06.

A few of our class poses at the sundial at the Lighthouse on San Salvador, Bahamas. See other beautiful phenomena from the Bahamas.

Tropical Field Courses -Western Program-Miami University

Biostatistical Analysis Demonstrated in Cushion Sea Star Study in Graham’s Harbor

The field experiment that was conducted relates to common scientific efforts in ecology and biology that are needed to produce statistical evidence to confirm a hypothesis. In our case, population ecology and environmental biology were a main focus. Estimating population is used when studying endangered wildlife and detecting detrimental effects to the environment in which we live in. Indicator species are noted for being able to gauge and monitor the well-being of a certain area where the population dwells. Environmental biology is incorporated with population ecology and can be used to estimate the difference and changes in a specified area by studying the organisms themselves. Our experiment on cushion sea stars Pteraster tesselatus used practical and simple ways of studying both population and the effects of the environment on the organisms themselves. The data that we have can be taken as evidence if the statistical reports are enough to stand up to the tests provided by ANOVA. In science, little can be proven without statistics and it is this integration of math that allows all subjects within biology to relate to one another.
The Experiment
Observations: Cushion Sea Stars have a preferable living environment. They are more densely populated in one area of the seabed than others.
Hypothesis 1: Sea stars will live preferably at a certain location (depth, distance from shore)
Hypothesis 2: Sea stars will thrive at this certain location and will grow optimally at this location. For the purposes of this experiment, size will directly correlate with optimal growth.
Hypothesis 3: Sea stars will prefer one substrate over another.
2 or 3 Tape Measures
Makeshift buoys (4 mesh dive bags, rocks, up to 4 snorkeling vests)
6 Snorkelers
2 driftwood stakes
Dive calipers
1 dive flag
A. Setting Guide Quadrant
-2 People act as “anchors on shore, measuring a 15 m stretch along the beach, setting driftwood stakes
-Anchors set tape measure at the stakes (each ‘Anchor’ anchoring their own tape measure)
-Anchor A sets tape measure limit at 20m while Anchor B sets his/her tape measure limit at 25m
-Swimmer takes the ends of both tape measures and swims out until both tape measures are taught, thus forming a 3-4-5 right triangle.
-At this point, the first guide buoy is dropped.
-Anchor A now sets tape measure limit at 25m and Anchor B sets his/her tape measure limit at 20m. Swimmer repeats process and drops second guide buoy
The guide quadrant is set
B. Quadrant Analysis
-Sea stars are counted and measured with dive calipers. Radius from mouth to tip of a leg is considered the “size” of the seastars.
-Depth at the beginning, middle, and end of the quadrant is measured. One reads the measurement while another dives down and holds tape measure at bottom.

C. Extending Quadrants
-2 swimmers swim out ~20 yards, guided by tape measures anchored at the already set dive buoys. They align the guide buoys to the stakes and set new dive buoys to mark the end of a new Quadrant.
-Procedure B is repeated until 4 Quadrants, extending from the shore, are analyzed
D. Extending Plots
-Procedures A to C are repeated until 5 total “transects” are analyzed and data is gathered.
Our experiment served several purposes in the study of cushion sea stars. We used a method called randomized area sampling of a specific population that can be used to approximate the entire population size. Random sampling methods are considered some of the most accurate among statistical methods in data collection. An example is the freshman year Frisbee lab, in which a Frisbee was thrown and all the clovers
(Trifolium sp.). underneath the Frisbee are counted. The Frisbee disc (approximately 30 square inches) is our sampling size. Divide the total area of the lawn (the supposed range of Trifolium sp.) by the size of the sample size. Then multiply that with the average number of Trifolium counted. Even a mathematical operation as simple as averaging is part of a system to evaluate numbers so that they are presentable in a supportive manner. When it comes to more mobile wildlife, such as avian or mammalian species, there has to be more complex and ingenious way of doing it. Trapping, tagging, and releasing animals is part a way to use proportion to create a logical (and surprisingly accurate) estimate of a total population. Recapture method involves capturing, tagging, releasing, and finally recapturing is a common way to use proportions to estimate a population. Traps are set out for a certain designated time. Each animal is tagged and recorded (usually size, weight, color, description, age, dental and health records, and many other aspects are recorded) and then released. A second set of traps is laid out in the same area after the first set has finished its course. Recapturing animals using traps will yield trapped animals that are both tagged and untagged. From this, we can presume that the ratio of tagged to non-tagged are representative of the ratio of a larger population. There are obviously certain flaws to this method. What if the animals that are trapped have been in contact with humans and domesticated to a degree? Or they simply prefer the bait used? Despite these misgivings, the randomization of the location of the traps and other safeguards help ensure that the numerical data that is collected is accurate.
The optimal growth factor in Pteraster tesselatus is defined as the ‘larger the better’: meaning that larger sea stars demonstrated optimal growth. However, optimal growth is not limited to just size and can be studied on a molecular level down to where optimal growth of a organ cell is determined by the abundance of a protein or another enzyme receptor. There are some that can be so severe that it leads to abnormal growth of certain cells. In a study of plasma membrane glycoprotein, contactinhibin, it was discovered that this density-dependent cell growth regulator has a profound effect on cell growth and the location of the cell in relation to this glycoprotein is a key element.
Although transformed cells express contactinhibin in a functionally active form, they are not growth-inhibited, suggesting that the defects that lead to their aberrant growth are located ‘downstream’ of contactinhibin.

Population Ecology is closely related to ecological studies. As a subgroup of ecology, population ecology relates closely to the study we conducted in Graham’s Harbor in San Salvador, Bahamas, specifically a branch of study called population dynamics that not only studies the numbers in a population, but the composition and make-up of different categories of individual members of that population (i.e. age, weigh, radius). Using this data, we can assess the well-being of a certain environment and possibly locate the source. In South America, harpy eagles were endangered and known to have been declining due to a variety of factors. By collecting data on their health, size, age, etc. in relation to an area, they were able to identify yet another factor in their declining population: the use of pesticides by local farmers. These pesticides thinned the wings of the birds and weakened their bone structure.

Our study measured sea star radius, which is one of many aspects that we could have measured. This was intended to allow us to show optimal growth at a particular depth: a sea star growing larger at medium depth will be taken as an indicator that sea stars grow more optimally at that depth than shallow depths and/or deeper depths. From observing the tides and environment, we can also deduce that the tides and waves disrupt sea star life in the shallower areas and that the micro-organisms the sea stars feed upon can not thrive well in too deep of water. Using statistical analysis (the numbers we collect) and relating it to the environment from which we gathered the data, we can come to at least tentative conclusions regarding question posed by our research team.
The advances in statistical studies are credited to Sir Ronald Fisher, who played a prominent role in the development of the ANOVA test: the F factor in the distribution of variables in ANOVA is named after this man. In the statistical analysis provided by our ANOVA, or analysis of variance, we are able to use our data as evidence in support of our claims: specifically the hypothesis stating that sea stars prefer a certain area over others. The most definitive data is showing that sea stars prefer a certain substrate over another. In all areas that were plotted, there was a drastic difference, but it should also be noted that it only compares only one variable: the substrate. A method that we used is known as the Tukey-Kramer method. This method uses a different confidence factor and deals primarily with the family error rate or the error rate of different sets of data. Little in the studies of science can there be proof without statistical evidence.
Without numbers, it would be merely a speculative affair. There are several tests that can be used to determine whether the statistics are viable to prove (or disprove) a hypothesis or idea. The tests we conducted are known as one-way ANOVA, because there is only one independent variable. Some of these tests are stricter than other and harder to pass. They help determine the level of accuracy of the data and to what degree a set of data can be used as evidence to prove something. For example, numerical studies on the populations of wildlife in Yellowstone National Park garner a lot of attention: without support from data collected, it would be a huge public embarrassment. Observations were made that herds of large game were booming in size. Correlating this trend with a decline in predatory animals eventually led to the reintroduction of certain predatory species, such as the wolf, into the park. If these had not been properly presented and supported, it is unlikely that it would have been taken seriously. Even though, this type of analysis and scientific work does not present itself in the public eye, it is necessary to convince experts of a field that such a problem is occurring.

For Further Info on this Topic, Check out this WWW Site:
Next Article
Previous Article
Return to Topic Menu

Here is a list of responses that have been posted to your discussion topic...

Important: Press the Browser Reload button to view the latest contribution.

If you would like to post a response to this topic, fill out this form completely...

Response Title:

Optional: For Further Info on this Topic, Check out this WWW Site:
Response Text:

DOWNLOAD the Paper Posting HTML Formating HELP SHEET!

We also have a GUIDE for depositing articles, images, data, etc in your research folders.

Article complete. Click HERE to return to the Pre-Course Presentation Outline and Paper Posting Menu. Or, you can return to the course syllabus

  • Tropical Marine Ecology of the Bahamas and Florida Keys
  • Tropical Ecosystems of Costa Rica
  • Site NAVIGATION--Table of Contents

    Listen to a "Voice Navigation" Intro! (Quicktime or MP3)

    Search WWW WITHIN-SITE Keyword Search!!



    Hays' Marine Ecology Images and Movies Ohio Bird Photo Collection | Tropical Bird Collection | Costa Rica Image Collection | Edge of the Farm Conservation Area | Hays' Tarantula Page | Local Watershed Fish Studies| Wildflowers, Arthropods, ETC in SW Ohio | Earth Science Resources | Astronomy Links | Global Change | Marine Ecology "Creature Study Guide" |


    | Educational Philosophy | Discovery Labs: Moon, Geologic Time, Sun, Taxonomy, Frisbee | Project Dragonfly | Vita |Field Course Postings | Student Research Postings | Nature/Science Autobiography | Environmental Programs at Miami University


    Daily Necessities: Macintosh Resources |Search Engines | Library Resources|Server Stats| Family Album | View My Schedule | View Guestbook | Western College "Multimedia Potpourri"

    It is 5:06:23 PM on Tuesday, October 20, 2020. Last Update: Wednesday, May 7, 2014