Bermuda Results

After the trip to Bermuda, all the recorded data was transferred into an excel spreadsheet so further analysis could be performed (Table 1). The two variables that were studied in this experiment were the level of predation and the temperature that the school of fish was exposed to in Whalebone Bay. In order to assess the impact that the predation level and temperature had on the school, three regression analysis calculations were performed (Table 2). The first regression analysis attempted to find a correlation between the temperatures of the water where the school was found with the average distance apart between individual fish in the school. The average distance apart was calculated by approximating the distance between individual fish by a number of fish lengths. The estimate was then multiplied by the average fish length in the school in centimeters. Ten fish were caught and measured and the average of the ten was used to determine a rough estimate of the distance between individual fish in the school. The results were also analyzed graphically to see if any patterns could be seen that the regression did not indicate (Graph 1).

Some fellow classmates entering the water at Whalebone Bay for a morning snorkel

Variable	F	df	P Value	R² Value
Temperature	0.513484	19	0.482829	0.027736
Predation	5.697764	9	0.044062	0.415963
Predation 2	10.76004	7	0.016817	0.642006

Table 2: The table above shows the significant values that were obtained from the regression analysis that was performed for each of the above variables against the average distance apart between individuals in the school. The predation variable includes all the data that was obtained. The predation 2 variable has the two data points which had trumpet fish as the main predators.

The regression analysis that was that compared the average distance apart with the total number of predators present yielded significant results. The P value of the analysis was 0.0441 and the F value was 5.69. This indicates that is a positive correlation between the average distance between individual fish in a school and the level of predation acting on that school. The results were also analyzed graphically to help determine the interaction between the two variables (Graph 2). Once the graph obtained and analyzed, it was noted that a few data points did not quite fit with the rest of the points. Upon further analysis, it was determined that the two points that did not fit the rest of the data, contained trumpet fish as the major predator. These two points were removed and an additional regression analysis was performed (Predation 2 in table 2 above). The results showed that there was now greater significance between the two variables. The F_1,7 value was 10.76 and the P value was 0.0168, indicating that the results without the trumpet fish values showed that there was a stronger correlation between the distance apart and the level of predation. The samples were also analyzed graphically and all of the points appeared reasonable (Graph 3).

As the week in Bermuda progressed, the size and the dynamic of the school also changed. The size of the overall school in Whalebone Bay decreased dramatically. It was estimated that the size of individual fish from the first day slowly decreased throughout the week and on the second to last day of the study it was roughly 10 times smaller than it was originally. On the last day of the study, the number of individuals increased significantly. It is also important to note that several hundred dwarf herring were present in the school for a few days in the middle of the week. The herring formed small, localized schools within the large silverside school.

Furthermore, the number of predators in the bay affected both the location of the school and their behavior. On days when the number of predators was lower, the fish spent more time closer to shore and under large mats of seaweed foraging. On the other hand, when the level of predation was higher, the fish spent more time avoiding predators. As the week progressed, the amount of seaweed dramatically increased and it provided both positive and negative effects.

A behavioral experiment was also performed in which I would swim through the school when there were different levels of predation present and observed the results. When there was a small distance between the fish and high level of predators, the fish would not be affected by my presence. They would constantly swim into my body and into my fins. One brave silverside even managed to swim into my snorkel. During the levels of low predation and greater distance between individuals, the school would keep a certain distance away. It also because more difficult to capture individuals using such a small net.