Ongoing Observations

The organism that I would like to study is the Wolf Willow (Elaeagnus commutata). I chose three areas that are along the Bow River in the southwest of Calgary, Alberta that have a large population of E. commutata. The first area (1) was between two pathways, and a medium distance from the river; the second area (2) was at a much higher elevation than the river, close to the river; and the third area (3) was at a higher elevation than the river and far away from the river.

The plants that were closer to the river in area 1 appeared to be taller, and still had some berries present on the tallest branches. I didn’t bring a measuring tape but the plants here are taller than I am, so they are at least 6’ tall. The plants in area 2 were around 4’ tall, and much more abundant than either of the other areas. Only a few scattered berries were present, again on the tallest branches. In area 3, the furthest from the river, the plants were quite short – only around 2’ – and there were no berries present.

It’s unusual to see berries on plants during winter, but the wolf-willow’s berries seem to remain throughout the winter and into the spring. They are a pale silver colour.

From these observations I hypothesize that the growth of E. commutatais heavily reliant on a nearby water source. I predict that the tallest E. commutataplants will be in close proximity to the river. The responding variable would be the average height of the plants, which is a continuous variable. The explanatory variable would be the adequate water present in the form of the Bow River.

Data Collection (#6)

Preparation for collecting data took almost as long as data collection itself. First, I made tables in Excel with which to collect my data, but ended up forgetting to include two rows for boreal climax species (birch and aspen). This was not a significant issue as I had included blank rows in the table which I used for this purpose. Then, using google’s random number generator, I determined which 6 blocks I was going to collect data from within. I then I used the random number generator again to determine the origin point of the transect axes. When this was finished I knew exactly where I was going to collect my data from before I went out. Finally, I made a 1x1m collapsible square out of thick cardboard and bolts to use as a frame to define my plots, much like the squares made of PVC tube and elbows in the video.

I was fortunate to have a friend come out with me and perform a lot of the recording as I called out my observations. Finding the blocks and starting points for the transects was occasionally difficult, but my gridded map and the GPS on our phones helped. We then strung a tape line down the transect axis and placed the plot square into the first position, recorded the relevant variables, measured the rose bush height with a measuring tape and calculated their average height, then moved the plot square to the next position. This was repeated until the transect was complete, at which point we found the starting point for our next transect and repeated the process. In total, 120 replicates were sampled over the course of 4 hours.

The main difficulty I encountered was determining which category certain variables fell within, especially about light and moisture. On several occasions I found myself wanting to assign a variable a value between two discrete categories by adding a .5. I did not do this. Aside from that, occasional game trails were encountered in the undisturbed areas, and I considered whether or not they may make an impact on the distribution and size of rose bushes. In the end I made a note of the game trails but did not alter the designation of the block they were found in.

My companion commented that solitary rose bushes seemed on average larger than those found in close proximity to others. I did not notice this pattern, but I will look for this when I analyse the data. If there is a discernible relationship, it may indicate intraspecific competition.

All in all, collecting the data was a relatively straight forward event due to planning it out ahead of time. No significant obstacles or set-backs were encountered.

Design Reflections (#5)

The acquisition of the initial data in module 3 was relatively uncomplicated, though I think I could improve the way I collect the data in a few mays. First by refining the categories; variables such as percentage of exposed soil and potentially relevant attributes of other species present were not recorded, and the method I used to categorize the moisture regime was imprecise. I may make changes to my tables to better isolate the relevant variables, and I will likely modify my technique to collect my data to allow for better numbers and accuracy.

The technique I used was to overlay a map of my chosen area with a grid of squares, each measuring 20×20, and designating each square as either disturbed or undisturbed. I selected one of each in close proximity and I divided each of these squares into a grid of 400 1x1m squares from which 3 were randomly selected in the disturbed square and 2 in the undisturbed square. These 5 plots were my samples, and I recorded the light exposure, moisture regime, and the other plant species present as my predictor variables, and the number and the average height of rose bushes present as my response variables. Because undisturbed squares outnumber disturbed squares, by selecting an equal number of each to collect an equal number of samples from, I hope to reduce the number of samples necessary to see relevant trends in the data.
Changes I will make to this technique will include the number of samples I record and the way I select them. I will still use randomization, but instead of recording observations from 2 and 3 randomly selected replicates per square, I will likely record 2 transects of 20 plots per 20x20m square, providing more data and an equal number of observations from disturbed and undisturbed areas. 

Sampling Strategies (#4)

Comparing the three sampling methods employed in the virtual forest tutorial, the haphazard method had the fastest estimated sampling time at 12 hours 26 minutes, then systematic at 12 hours 36 minutes, and finally random with the slowest estimated sampling time at 12 hours 43 minutes.

For the two most common species (Eastern Hemlock and Red Maple), the systematic method was the most accurate, with p = 1.30% and 12.53%. The least accurate was the haphazard method (18.81% and 19.43%), with random falling between the two (12.26% and 15.90%).

For the two least common species (White Pine and Striped Maple), the systematic method was still the most accurate (42.86% and 37.14%) the haphazard was still the least accurate (98.81% and 114.29%), and the random method still fell in the middle (50.00% and 100.00%).

The accuracy of all methods declined along with species abundance, and the systematic approach remained the most accurate regardless of abundance.

Ongoing Field Observations (#3)

I have chosen Rosa acicularis (prickly rose) as the organism I will study. It is common throughout the southern Yukon and can be found easily in most areas that have not been heavily developed. Around pumphouse pond, the area I have chosen to conduct my observations, I have observed this rose in various densities and morphologies. It appears most frequently along the sides of walking trails, power-line clearings, cleared lots, and at the edges between undisturbed and developed sites. It appears infrequently in the midst of stands of tall coniferous trees, far from trails and other disturbances.
Almost everywhere I found Rosa Acicularis growing, I also found Rhododendron groenlandicum (Labrador tea), though the inverse was not true. In fact, in many of the darkest, wettest parts of the forest, where prickly rose was least abundant, Labrador tea had grown leaves and appeared to be further in its seasonal development than in areas where the prickly rose was more abundant, where Labrador tea was still bare.

One area in particular, on the top of a bank at a clearing to the east of pumphouse pond, had a thicker-than-usual stand of rose bushes, most of which were above average in height, interspersed with fire weed (Chamaenerion angustifolium), a known pioneer species.

My hypothesis is that Rosa acacia is also a pioneer species which tolerates and/or requires many of the conditions present early on during a secondary succession. I predict that a random sample of plots representing gradients which include regimes of light, moisture, and inter-species interactions, will reveal discrepancies in the size and abundance of rose bushes.

Potential predictor variables include: light exposure, moisture regime, and other species present (# and type). Response variables include rose plant abundance, average plant height, and average fruit density. I am curious to see if, with sufficient replicate plots, any of these predictor and response variables can be specifically related to each other. For the sake of simplicity, all three predictor variables will be categorical, and all response variables will be continuous.
I will continue to use pumphouse pond as a general study location, though I will apply a grid over a map of the area and divide it into blocks which contain disturbances (roads, trails, developments) and blocks which do not, and randomly select an equal number of each category to sample 10 or more randomly selected sub-plots with respect to the predictor and response variables listed above. 

Blog Post 2: Sources of Scientific Information (Percy)

The source of scientific information is referenced below:

Ding, Y., & Peng, J. (2018). Impacts of Urbanization of Mountainous Areas on Resources and Environment: Based on Ecological Footprint Model. Sustainability,10(3), 765.

Based on what I have learned on how to evaluate sources of scientific information, I have determined that this article is peer-reviewed, academic review material. The information was written by two authors who have knowledge on the subject, as Jian Peng is the one who submitted the data analysis and it was reviwed by Yu Ding who wrote the paper and handled the manuscript submissions. The paper had been reviewed by at least one referee before publication, as we can see the submission dates and the publication dates are different. The source also reports results of an experiment by including a methods, results, and discussion section. The research is supported by in-text citations, and followed by a bibliography at the end of the paper.

Blog Post 6

Collecting the data for my study has so far gone pretty smoothly. It has been a little challenging to lay the sampling line as the vegetation is fairly dense and quite a bit of scrambling has been required. I have sampled 6x 15m lengths at each site, and on each sampling line I have taken 5 measurements of fern frond size. As I began sampling next to the creek site, I realised that the fern growth became more abundant as the distance from the path increased, so I made sure that I always recorded the site closest to the road as sample 1 and the one farthest from the road as site 5. From my initial observations, it appears that the fern size variability seems greater at the creek site compared to the interior site, however, this is yet to be determined if it is a statistical difference.

Blog Post 5

I modified my data collection strategy from my initial plan, as I could not do spore counts on the fronds as it is spring time and the new fronds did not have spores and the mature fronds had dispersed their spores. Instead I measured the width of the widest pinna of the longest frond, which still gives an indication of the capacity of the frond to carry spores, so I think this approach will work. I also decided to just count the new fronds per crown as it was very difficult to count all the fronds, as many were rotting or broken and it was hard to determine which were viable fronds and which were not. Just counting the new fronds should still give an indication of the vitality of the fern.

In terms of laying out a 15m rope marked at 3 m intervals and then selecting the closest fern to that point, this worked well and I will continue to do this. I will increase the distance between sampling lines from 1m to 2m as the ferns can be as much as 2m across so I do not want to be counting the same fern on two different sampling lines.

Post 4

Comparing the three different sampling strategies, the fastest sampling time was with the haphazard method (12 hours 30 minutes), then systematic random (12 hours 36 minutes), and then random (12 hours 42 minutes).

When comparing the percentage errors using the different sampling techniques for the two most common species, the most accurate method was systematic random sampling (7.2% and 5.5%). The next lowest sampling percentage error was with the haphazard technique (10% and 13.4%) and the highest sampling percentage errors was with the random sampling technique (11.5% and 25.5%).

When comparing the percentage errors using the different sampling techniques for the two rarest species, the most accurate method on average was haphazard sampling (52.6% and 48.8%). The next lowest sampling percentage error was with the systematic random technique (5.1% and 170%) and the highest sampling percentage error was with the random sampling technique (52.6% and 248%).

The accuracy with all techniques was greatest with the most abundant species. When the species were abundant the most accurate technique was the systematic sampling technique, but when the species were rare, all the methods had high error rates. It is difficult to conclude that one sampling method is superior to others as there was a wide range of species abundance, so I would select the method that required less time to complete.

Blog 3

The organism I have chosen to study is the Western Sword Fern (Polystichum munitum), a common fern found in the Pacific North West region.  The fern is commonly found in moist, mild and shady environments.

The reason I decided to study this organism is that it is commonly found throughout Pacific Spirit Regional Park, however there were some locations on within the gradient of study that I observed fewer sword ferns present. At the creek site (Site 2), on the west bank of the stream there were fewer ferns, however on the east bank of the stream there were many more. The ferns at the interior site (Site 1) appeared to be more vigorous and larger than the ferns at site 2. Site 1 is located amongst large coniferous trees and it appears that less light makes it through to the forest floor. This is an environment that would suit sword fern growth so I would predict that the ferns may grow more vigorously or have greater abundance at this site compared to the creek site (site 1). The banks of the creek are relatively open to the sun as the creek runs south and the trees at the edge of the forest have been thinned or cleared as the creek passes beneath the road. I would expect that this site, although moist, the sun exposure may affect the sword fern growth. I did notice that the ferns were more prevalent on the east side of the creek than the west side, and again this may be due to the sun exposure the different banks experience.

I hypothesize that the fern growth will be greater in the more shady areas of the forest as measured by crown count per unit area, and frond length. I am also interested in counting the density of spores to determine if that also varies by site. The explanatory variable will be the amount of light at each site classified as high and low. The response variable of crown number/ plot and frond length will also be continuous variables.

Sample of field journal entry including observations of different sites.