Pre-Lab
What are the variables in this experiment?
Independent - the amount of fish that you tag and the number of catches of fish that you take
Dependent - the accuracy of the popultion size
Controlled Variables - same paper bag, same person grabbing the catches, same size of beads
Control Group - the size of the fish and the size of the bags
Experimental Group - the different population numbers and the number of catches taken
Independent - the amount of fish that you tag and the number of catches of fish that you take
Dependent - the accuracy of the popultion size
Controlled Variables - same paper bag, same person grabbing the catches, same size of beads
Control Group - the size of the fish and the size of the bags
Experimental Group - the different population numbers and the number of catches taken
Problem
How accurate is the catch and release method when trying to determine the size of a population?
Hypothesis
If different numbers of tagged fish and number of catches are taken, than the group with the largest amount of tagged fish and number of catches taken will have the greatest accuracy of determining the population size.
Materials
· Paper Bag
· Gold Beads
· White Beads
· Pencil
· Paper
· Calculator
· Gold Beads
· White Beads
· Pencil
· Paper
· Calculator
Procedure
1. Pick a random bag
2. Read off the number on your specific bag
3. Look on the projector to look up how many fish you will be tagging and how many catches you will be doing
4. Replace the respective amount of white beads from the bag with gold “tagged fish” beads.
5. Record the original number tagged into the table.
6. Shake the bag and mix the “tagged” beads with the original beads.
7. Without looking into the bag, grab a handful of beads and record the total amount and the amount of “tagged” fish.
8. Repeat this for the total of the assigned catch size.
9. Calculate the percent tagged for each catch and average the column.
10. Calculate the estimated population size by dividing the number tagged by the mean of the sample percentages.
11. Count the actual population.
12. Calculate the percentage error between the estimated population and the actual population.
13. Record the data into the class set on the projector
2. Read off the number on your specific bag
3. Look on the projector to look up how many fish you will be tagging and how many catches you will be doing
4. Replace the respective amount of white beads from the bag with gold “tagged fish” beads.
5. Record the original number tagged into the table.
6. Shake the bag and mix the “tagged” beads with the original beads.
7. Without looking into the bag, grab a handful of beads and record the total amount and the amount of “tagged” fish.
8. Repeat this for the total of the assigned catch size.
9. Calculate the percent tagged for each catch and average the column.
10. Calculate the estimated population size by dividing the number tagged by the mean of the sample percentages.
11. Count the actual population.
12. Calculate the percentage error between the estimated population and the actual population.
13. Record the data into the class set on the projector
Observations
The largest tagged fish size was 30 fish and the second largest amount of catches taken was 15. One group that had 30 fish tagged and 15 catches had a percent error of 124% and the other one had a percent error of 2.5%.
The one group with the largest amount of tagged fish and the largest amount of catches had a percent error of 32.5%
The group with the least percent error (.5%) had 20 fish tagged and 20 catches
The one group with the largest amount of tagged fish and the largest amount of catches had a percent error of 32.5%
The group with the least percent error (.5%) had 20 fish tagged and 20 catches
Analysis
1. What could cause the population size estimate to be different from the actual population size?
Estimates are never going to be exactly the same as the actual population for a number of factors. For one, it’s impossible to be able to count every single individual of the population so the two numbers are naturally going to be different. Another reason is that when the tagged population is caught, if by random chance you get many of the tagged species in one catch or none at all it significantly changes the estimate.
2. How can the size of your catch sample compared to the actual size of the population affect population estimate?
My specific group had 20 tagged fish and we took 10 catches. With the results from our catches, we thought that our population was 266, but, in reality, the population was 229. We had a 16.2% error. This isn't a huge percent error to have; however, I wouldn't exactly call this a reliable way to count the population of fish considering that we were 37 fish off.
3. How can the number of samples (or catches) affect the results?
From the data, it appears that there is no real pattern in the number of catches and the percent error. The high percentages of error we experienced could be explained by us not pulling a tagged fish in many of the trials. Not pulling a tagged fish significantly alters the results and this could easily happen if we took very small samples each time from a large population of fish. I think that taking large catches would actually give a much more accurate figure. The more samples that are done, you have a more accurate figure because you have a better average. Because we were allowed to take as small or as big of a handful of fish as we wanted, I think our results suffered.
4. Does this method appear to be an effective way to assess population size? Why of why not?
No, this does not appear to be an effective way to assess population size at all. The data collected shows absolutely no correlation. For every combination of sample size and number of samples one group would turn in their data with a high percentage of error, but then another group with the same combination would have a low percentage of error. I believe this is due to just random chance when picking the tagged fish. If one group happens to not pull a tagged fish a few times their data is going to be much different than a group that pulled a tagged fish every time. Part of this problem could be due to groups that pulled a very small amount of fish for each catch.
5. If you were predicting a large population (like an actual ocean fishery) would your percent error be very large?
Yes, I think the percent error would be very large. An ocean fishery has so many fish that the tagged fish are very easy to not be caught and therefore creating a large percent error.
6. What concerns should a biologist have about a species' habits before (s)he uses this method to approximate the size of a population?
One concern is the species behavior with other individuals of the same species. Many species either like to group together such as fish, or others isolate themselves off and do not interact with other individuals of their same species. This could make it very difficult to even find the species and catch a large sample species. Another habit that would be difficult to tag is species that avoid human confrontation. Many species of apes will flee humans and that makes it very difficult to tag, and get any sample size at all.
Estimates are never going to be exactly the same as the actual population for a number of factors. For one, it’s impossible to be able to count every single individual of the population so the two numbers are naturally going to be different. Another reason is that when the tagged population is caught, if by random chance you get many of the tagged species in one catch or none at all it significantly changes the estimate.
2. How can the size of your catch sample compared to the actual size of the population affect population estimate?
My specific group had 20 tagged fish and we took 10 catches. With the results from our catches, we thought that our population was 266, but, in reality, the population was 229. We had a 16.2% error. This isn't a huge percent error to have; however, I wouldn't exactly call this a reliable way to count the population of fish considering that we were 37 fish off.
3. How can the number of samples (or catches) affect the results?
From the data, it appears that there is no real pattern in the number of catches and the percent error. The high percentages of error we experienced could be explained by us not pulling a tagged fish in many of the trials. Not pulling a tagged fish significantly alters the results and this could easily happen if we took very small samples each time from a large population of fish. I think that taking large catches would actually give a much more accurate figure. The more samples that are done, you have a more accurate figure because you have a better average. Because we were allowed to take as small or as big of a handful of fish as we wanted, I think our results suffered.
4. Does this method appear to be an effective way to assess population size? Why of why not?
No, this does not appear to be an effective way to assess population size at all. The data collected shows absolutely no correlation. For every combination of sample size and number of samples one group would turn in their data with a high percentage of error, but then another group with the same combination would have a low percentage of error. I believe this is due to just random chance when picking the tagged fish. If one group happens to not pull a tagged fish a few times their data is going to be much different than a group that pulled a tagged fish every time. Part of this problem could be due to groups that pulled a very small amount of fish for each catch.
5. If you were predicting a large population (like an actual ocean fishery) would your percent error be very large?
Yes, I think the percent error would be very large. An ocean fishery has so many fish that the tagged fish are very easy to not be caught and therefore creating a large percent error.
6. What concerns should a biologist have about a species' habits before (s)he uses this method to approximate the size of a population?
One concern is the species behavior with other individuals of the same species. Many species either like to group together such as fish, or others isolate themselves off and do not interact with other individuals of their same species. This could make it very difficult to even find the species and catch a large sample species. Another habit that would be difficult to tag is species that avoid human confrontation. Many species of apes will flee humans and that makes it very difficult to tag, and get any sample size at all.
Class Data
Conclusion/Real-life Example
According to the experiment, my hypothesis turned out to be incorrect. I thought that a greater catch size and more samples would have a more accurate percentage of error because it’s sampling more of the population and there are more trials to get a more accurate figure. The experiments results proved that there is actually no correlation between the number sampled and the number of samples. Every combination of catch sizes and number of trials all yielded some high and low percentages of error. This is because this experiment is so easily affected by chance of not catching any tagged fish. If you happen to not catch any tagged species, the percentage of error is significantly changed and the data is much less accurate.
To make this experiment better, more guidelines are crucial. You cannot just take a very small population of fish during each catch; however, you need to keep the catch sizes random to keep to the method of using catch and release. Therefore, the experiment as a whole is inaccurate. People who made small catch sizes usually didn't grab any tagged fish in their catches and this created many zeros, which threw our population estimation way too.
The result of this experiment makes current populations estimates very questionable. It was evident from this experiment that using tagged species to determine population size is extremely inaccurate with one trial having a huge percent error of 124%. If this is the case, all the population estimates of animals such as endangered species make these estimates much less credible. For example, the population of pandas, an endangered species. Pandas are very illusive in nature so even getting that species tagged is a difficult task. Then these tagged species can, by chance, avoid being caught in the trials and thus making the population of pandas seem much lower than they actually are. In my opinion, save the time. It is not worth the work to go around tagging pandas, or any animal in that respect, only to get inaccurate data.
To make this experiment better, more guidelines are crucial. You cannot just take a very small population of fish during each catch; however, you need to keep the catch sizes random to keep to the method of using catch and release. Therefore, the experiment as a whole is inaccurate. People who made small catch sizes usually didn't grab any tagged fish in their catches and this created many zeros, which threw our population estimation way too.
The result of this experiment makes current populations estimates very questionable. It was evident from this experiment that using tagged species to determine population size is extremely inaccurate with one trial having a huge percent error of 124%. If this is the case, all the population estimates of animals such as endangered species make these estimates much less credible. For example, the population of pandas, an endangered species. Pandas are very illusive in nature so even getting that species tagged is a difficult task. Then these tagged species can, by chance, avoid being caught in the trials and thus making the population of pandas seem much lower than they actually are. In my opinion, save the time. It is not worth the work to go around tagging pandas, or any animal in that respect, only to get inaccurate data.
Works Cited
"Endangered Species." Endangered Species - EndangeredSpecie.com. N.p., n.d. Web. 18 Mar. 2013.
http://www.endangeredspecie.com/.
http://www.endangeredspecie.com/.