https://docs.google.com/document/d/1B1UZbwHUGQDY37oAWqlhUOju5m0BSuSu0HJFnOlT2Sc/edit
What you’re doing:
This activity will teach you a commonly used method to estimate population sizes of mobile animals called Distance Sampling. This technique is based on the assumption that a researcher cannot observe every individual of a population, but is more likely to see those close to them than those far away. Typically a researcher will travel along a predefined transect line and will record individuals seen and their estimated distance from the transect line. At the end of the survey, the number of individuals that are likely within the area but remained unseen are estimated using an algorithm and used to calculate a population density. Researchers can then multiply that density by the total study area to estimate their population size. This technique could be used for either marine mammals or seabirds by running boat surveys over transect lines.
Running The Simulation:
Simulation controls:
Use the mouse to control the camera
Use the WASD or the arrow keys to move the boat
Use the Z key on your keyboard to toggle zooming in and out
In this simulation, you will act as a researcher studying humpback whales in the Gulf of Maine. You will first be trained on species field identification and distance estimation. In the field, whales are usually identified to species using a variety of surface cues (things you might see when the whale arrives at the surface to breathe). If the whale is far in the distance, all you may see is the spout or blow of that whale- the condensation that occurs when the warm breath leaving their blowhole interacts with the cooler atmosphere. Trained observers can actually identify whales to species using just the spout shape and size. If the animal is closer to the vessel, an observer may see the actual body of the whale and use those visual cues (size, shape of dorsal fin, fluke, etc.) to identify the species. Estimating distance also has a steep learning curve! Researchers will often train themselves using technology such as laser rangefinders, but will often rely on their own estimates or binocular reticlesLinks to an external site. to record distances of the animals.
Instructions:
Once you have completed training, you will run the actual research survey simulation in Assessment Mode to estimate the population size of the humpback whales. As you work through the “Assessment” simulation, search along the transect line for humpback whales. On a notebook, record distances of each sighted whale (humpbacks only!) from the transect line. Be sure to also record distance traveled (10 km is the max if you make it to the end of the transect line.) When completed, follow the directions to calculate your estimated population size. Enter your population estimate (calculated below) into the Assessment simulation. Take a screenshot of your population size entered in the box before you submit and take a screenshot with the actual population as reported by the simulation.
Calculation:
When you complete your sampling in the assessment mode, you will try to estimate the actual population size using the instructions and equation provided in the assignment. Below you will find an example of the calculation.
Plugging your data into the equation will give you the density of whales in the area. You will then multiply that density by the total area to estimate how many animals are likely out there.
Density Equation: (n+u) / 2wL
n= number of whales sighted by observer
u= number of whales in the area undetected
w= width of sampling (found by the farthest detected whale)
L= transect length traveled
Example data from a survey can be found in this spreadsheetLinks to an external site.. In this example, the user saw a total of 18 humpback whales and recorded the distance of each sighting. This serves as “n”.
We assume there are still more whales in the region that were not detected. For the density equation, you need to know how many whales you did not detect. This is determined by creating a histogram and counting how many whales should be sighted at farther distances. In this example, u is 32 whales. We assume that even though whales were distributed evenly throughout the landscape, sightings became less frequent as whales were farther from the ship. This means the number of whales in each distance bracket should be the same. For example, in the example histogram, the observer saw 10 whales in the closest bracket. Each bracket should have had 10 whales seen as well (but were not observed because they were harder to see). Count the unseen whales within the area and call this “u”, which, again, in this example is 32. In your histogram, identify the highest number of sightings and assume each bracket should have that amount. In this examples, L is 10 km and w is 1.65 km. Note: If you did not travel the entire line (i.e. you stopped sampling early), use your final distance traveled rather than 10 km and use the farthest detected whale as your width.
Your calculation would then look as follows: (18+32) / (2*1.65*10) = 1.5
This means you would expect 1.5 whales per square kilometer! Given our sample area is 100 square kilometers, we would multiply 1.5 by 100 to get our final estimation of 150 whales in the region.
Using your observation notes, create a histogramLinks to an external site. of your distances, either in a computer spreadsheet or on paper. Your histogram should represent how many whales were sighted at each distance range.
Count up the total number of whales you saw
Calculate the undetected whales using the above method
Enter the n and u values, as well as the width and transect length into the equation
Calculate the population density
Interpretation:
It is important to realize the number that you get is an estimate and not the actual number of whales out there. There are many factors that can skew your calculations to either overestimate or underestimate. These may include how accurately you estimate distances, if you missed animals that were very close to the boat (as we assume you will detect all of those whales), the distribution of animals over the area, how far you traveled over the area, the speed of the boat, etc. This can happen on a real survey, but especially in our virtual ocean (which is still a work in progress).
Most of your estimates will likely be significantly higher than the actual whales reported at the end of the game. This method assumes that animals are equally distributed over the landscape, meaning the same number of whales would be close to the boat as you would find far away from the boat (i.e. animals are not clumped together). In previous versions of the simulation, students were having difficulty finding whales in the assessment mode. To combat that problem, programmers brought more humpback whales close to the boat so that they could be easily found. This then creates a bias leading to an overestimation (the density of whales is not nearly as high at farther distances as it is at close distances).
It is 100% fine if your estimation is very different from the reported whales. The major goal of this assignment is to be exposed to the challenges that exist in collecting and reporting field data on offshore animals. As long as you walk through the steps to the best of your ability, you will get full credit for this assignment!
Reflection
As an assessment, write a reflection describing your experiences. In your written reflection:
List the population size you calculated for whales (include a screenshot of your activity output) and show your calculations
Describe factors you think contributed to the accuracy (or inaccuracy) of your population estimate in comparison to the actual number of whales in the simulation. Include the actual population size in your comparison.
Speculate on how you could improve your population estimate if you ran the simulation again.
Describe biases this technique may have or errors that may occur.
Describe how you might alter this type of survey in studying seabirds.
Rubric
Week 4 Assignment I: Virtual Ocean Distance Sampling (1)
Week 4 Assignment I: Virtual Ocean Distance Sampling (1)
Criteria Ratings Pts
This criterion is linked to a Learning OutcomeCriteria #1 Population Estimate
Estimate reflects application of research techniques from the field environment. Includes population density estimate and calculations, as well as actual population value.
40 pts
This criterion is linked to a Learning OutcomeCriteria #2 Reflection Content
Writing demonstrates self-reflection on research performance and describes specific factors that may have caused overestimates or underestimates. Reference is made to learning materials. Methods are appropriately altered to be taxa-specific for seabirds.
40 pts
This criterion is linked to a Learning OutcomeSources
Outside sources relevant to the discussion are utilized and cited using APA style.
10 pts
This criterion is linked to a Learning OutcomeConventions of Writing
Uses correct grammar, appropriate terminology, and spelling.
10 pts
Total Points: 100
