For the initial data collection I used the random sampling strategy in a 1m x 1m area. I chose this technique not only because I thought it would be the most efficient but also, for the soil samples I didn’t want any bias involved as to where they were taken from. The only difficulty I can across was samples in the exact areas where for example a sagebrush plant or fir tree was. In this case I just took the soil sample directly beside the plant or tree. I will continue to use this technique to collect my final data.
Category: Percy Hebert
Post 4: Sampling Strategies
The virtual forest tutorial allowed for the testing of three different sampling strategies: systematic sampling, random sampling and haphazard sampling. It was determined that the fastest sampling time was for systematic sampling at 12 hours, 34 minutes. Followed by haphazard sampling and random sampling, 12 hours and 37 minutes, and 12 hours and 47 minutes, respectively.
The two most common tree species were the Eastern Hemlock and Red Maple. The most accurate of the three techniques for the Eastern Hemlock was systematic sampling with a 1.3% percent error, and the most accurate for the Red Maple was Haphazard with a percent error of 5.4%. The least effective sampling strategy for Eastern Hemlock was Haphazard with a percent error of 45.4%, and the least effective for Red Maple was random sampling with a percent error of 68.4%. Additionally, systematic sampling had a percent error or 26.0% for Red Maple, and random sampling had a percent error or 7.3% for Eastern Hemlock.
The two least common tree species were the White Pine and the Striped Maple. Systematic sampling was the best technique for White Pine and the worst technique for Striped Maple, with percent errors of 4.8% and 174.3%, respectively. The next best sampling strategy for White pine was haphazard sampling with a percent error of 98.8%, followed by random sampling with a percent error of 147.6%. For Striped Maple, the best sampling strategy was random sampling with a percent error of 18.9%, followed by haphazard sampling with a percent error of 66.9%.
Overall, the accuracy of the sampling got worse with lower species abundance as shown with the difference with the two groups above, the most common and least common species. For example with a larger species abundance (the most common species group), the percent error did not go above 68.4%. However, with a smaller species abundance (the least common species group), the percent error went as high as 174.3%. Overall, there is not one sampling strategy that clearly stands out as the most accurate, however, Systematic sampling is the closest, it has the two lowest percent errors for Eastern Hemlock and White Pine. It is hard to tell which is more accurate between random sampling and haphazard sampling as their percent errors are all fairly similar.
Post 5: Design Reflections
I did my initial data collection using the distance haphazard sampling method. I divided the accessible areas at Pipers Lagoon into 6 areas: A through F. I began with only Area A, where I designated five points spread out over the area. I found the sampling technique to be fairly straight forward. Often I found myself confused by my compass which took several seconds to find North. When I indicated the closest tree type in each quadrant, some were of almost equal distance which caused me to refine my guidelines and choose points which created no bias decisions.
The data I collected slightly surprising, as one tree type, the red alder was nowhere in the area, as well as Douglas fir in which a single young tree was recorded. Outside of the data collection I noticed neither of these trees were present. Otherwise, the relative abundances of Garry Oak and Arbutus trees appeared to be representative in which the Garry Oak was slightly dominant.
I believe this sampling method was sufficient in identifying the tree type and relative abundance. Initially, I created a table in which I would write young, mature, very mature, or dead to describe the tree. I believe using a table with a different layout will be easier where I can tick the box which applies rather than having to write down a letter.
By modifying the data collection table I can speed up the sampling process and increase the amount of area covered per hour. Therefore, the research can be done more continuously in a single day rather than requiring several days.
Blog Post 8
Blog Post 8: Tables and Graphs
Create a blog post discussing your table or graph. Did you have any difficulties organizing, aggregating or summarizing your data? Was the outcome as you expected? Did your data reveal anything unexpected or give you any ideas for further exploration?
As my data set was small there was not a need for any graphs such as a boxplot or scatter plots. I used pie charts and bar graphs to show the relative frequency and relative densities calculated in my study- to show an overall picture of what tree patterns were on each substrate. I did have difficulty with the calculations at first as they all seemed foreign to me. At the end of the calculations, I had a bunch of numbers that I was overwhelmed with, trying to figure out how I could tell the story I was trying to with all of the decimal places. I found myself wanting to make graphs more complicated then they needed to be. To look for data that may show something more then the simple story I was telling ( jack pine are in higher frequency and density amongst bedrock). Stepping back from trying to compare myself to the reports we have been reading in class, and looking at what my project actually was – a simple observational study helped give me some clarity. This helped in simply doing the calculations and making the bar graphs without complicating the paper with unneeded graphs and tables. Though I do hope that I represented the data to its fullest. I could have potentially done more with the DBH data.
The outcome of the experiment was what I expected. It would be interesting to measure distances between jack pine populations on different bedrock areas to see if there is a preferred spacing to avoid competition with each other.
Blog post 9
Blog Post 9: Field Research Reflections
Create a final blog post that reflects on your field research. You both designed a field experiment and then carried it out. Did you have any issues with the implementation or have to make any changes to your design? Has engaging in the practice of ecology altered your appreciation for how ecological theory is developed?
Carrying out the field project from design to implementation was a fairly uncomplicated process. The POC technique that I used is a well studied and easy to follow strategy for quantifying vegetation for a large area of land. Next field project I think that I would gather more samples for the data to be more representative of the population. I also would have liked to apply a few more statistical values to show proper comparisons of two separate populations. The steep learning curve for me came in writing the report. The process of figuring out excel and learning the language of ecology and then trying to apply it to my research took a lot of brain power. It made me realize how important it is for learning to have a community of like-minded individuals around to bounce ideas off of and troubleshoot problems one may be having. Learning how to implement a design and write a report was eye-opening and hopefully rewarding ( we will see how the grading process goes) but I feel my appreciation for the practice of ecology would have been more conducive if there was an easy to access community to communicate with.
Post 4: Sampling Strategies
For my Virtual Forest sampling strategies I chose the Snyder-Middleswarth Natural Area. The fastest sampling method was systematic sampling requiring only 10 hours of sampling while random and haphazard sampling required ~12.5 hours each. This appears mainly due to the amount of travel time to and from sample locations.
Common species such as Eastern Hemlock, Sweet Birch, and Red Maple appeared to be best represented by random sampling and haphazard sampling. Both produced low and sometimes negative values, indicating the true value and estimated value were extremely similar, or slightly overestimating.
More rare species of lower abundance, such as Yellow Birch, Chestnut Oa, Striped Maple and White Pine were often over represented through sampling. The systematic sampling indicated no Striped Maple or White Pine at all in the area which completely removes species from the area analysis. The haphazard sampling method demonstrated lower percent errors indicating that it is the most useful while analyzing rare species. The accuracy of species density decreased when analyzing rare species, % errors or over 100 or -100 indicate the data is not representative.
The sample size of 24 functioned well for random and haphazard methods, however, with the systematic method completely missed two species types. These rare species may be included in the data more efficiently if the sample size was increased to 50. Therefore, 24 was not sufficient in estimating the total abundance of tree species.
Post 3: Ongoing Field Observations
1. Identify the organism or biological attribute
I have decided to study the occurrence of four tree species at Pipers Lagoon Park; Quercus garryana, Arbutus menziesii, Pseudotsuga menziesii, and Alnus Rubra.
2. Observations of organism or biological attribute along an environmental gradient.
I decided to observe the relative abundance and occurrences of species at different locations around the park. In my field notes, I compared four physically different areas which all had different combinations of surrounding influences such as their proximity to water, orientation to north, slope angle, growing substrate, and surrounding vegetation.
I found that Alnus rubra was least abundant throughout the park, and all that were present were extremely young. Young Arbutus menziesii were growing between dense Pseudotsuga menziesii trees. There were vast amounts of both Quercus garryana and Pseudotsuga menziesii trees, where Quercus garryana appeared to occur mainly on rocky coastlines closes to the water.
3. Postulate one hypothesis and make one formal prediction based on that hypothesis.
A major environmental influence observed while traversing the area is direct and altered wind. Along the Northern coastline waves and wind hit the land vigorously and create a relatively more harsh environment. The South portion is protected by physical effects from the lagoon, where waves are dispersed and the area is protected from wind by the length of the tied island.
I have also considered the amount of sunlight each portion of the island receives. On the North side the sun rises without being blocked, however, on the South side the sun becomes hidden by a hill diminishing the amount of net sunlight possible.
Hypothesis: Quercus garryana trees are more abundant where there is a large amount of sunlight and protection from direct ocean wind.
I predict that Quercus garryana will be most abundant on the South coastline where there is little to no ocean breeze and a gradual landscape. I also predict that Pseudotsuga menziesii will be most abundant on the Northern coast where there is harsh wind and rocky bluffs.
4. Potential response variable and potential explanatory variable and whether they would be categorical or continuous.
The potential response variable is tree species abundance, which is categorical in which the trees will be assigned to their respected category. The predictor variable is relative exposure to ocean breeze/wind, which will be categorical relative to each site conditions.
The experiment is not manipulative, and is defined as a logistic regression experimental design for analyzing the presence or absence of four tree species due to environmental influences.
Blog 9: Field Research Reflections
Over the course of my field study, there were a few times where I could have changed methods or revised my overall hypothesis but chose to be stubborn and stick it out, and at other times, the feedback I received from others helped my revise my study in a few ways. The support that I received from my parents and sister and their help with the study was very important to me and it was amazing to have help when I needed them and I could not have done my study properly without their help. Ecology was an unknown world of information to me before I took this course and I never imagined in the beginning how much of the information would pertain to the real world, as some other courses don’t always do. I also definitely did not think I would learn so much about the Mallard duck and their habitat preferences. At times I struggled with managing my time, especially as my other courses started at TRU in September and I had originally wanted to be finished by the time the fall semester started but things don’t always go as planned. Something that definitely would have helped would have been reading some literature related to my study closer to the beginning of the course which may have helped guide my methods of collecting data rather than leaving it to the end of my study. Taking part in this course has increased my knowledge and appreciation for how scientists hypothesize ecological theories and how they conduct their research!
Blog Post 8: Tables and Graphs.
The graph above shows a comparison of the average alkalinity from four water samples collected at each of the three locations on McArthur Island in Kamloops, B.C. This is a sample as I have not yet figured out how to properly insert the standard deviation bars but it will be added on as a graph element for the final report.
From my graph, it is quite obvious to see that the pond had much higher alkalinity which ducks tend to like, as stated in most of the research I read through. Though my observations and data collections show that very few ducks tended to be in the pond compared to the other two locations. I hypothesized that ducks prefer the moat by the bridge over the pond or the moat entrance to the Thompson River and from my duck counts over the 4 days of my collection, the hypothesis was supported. Though, since the water is close to neutral pH and is lower in alkalinity, and so far, the portion of my study that compared the number of ducks in the shade versus the sun doesn’t seem to show a pattern, it seems to suggest that there are other reasons why the ducks prefer the bridge area.
Post 2: Sources of Scientific Information
The article Estimating Carrying Capacity for Sea Otters in British Columbia published in the Journal of Wildlife Management is an example of academic peer-reviewed research material. 
The authors are professionals in their fields, associated with environmental consulting, the department of Fisheries and Oceans, and Malaspina University-College.The article contains in-text citations and appropriate references of citations. The authors include their methods and results of their study. Although dates of submissions and edits are not noted, the acknowledgements reference two anonymous reviews.
Gregr, E., Nichol, L., Watson, J., Ford, J., & Ellis, G. (2008). Estimating Carrying Capacity for Sea Otters in British Columbia. The Journal of Wildlife Management, 72(2), 382-388. Retrieved from http://www.jstor.org.ezproxy.tru.ca/stable/25097550
