Recent Posts

Blog Post 1: Observations

TRU Open Learning Leave a comment

User:  | Open Learning Faculty Member: 

The study area I have chosen is Central Park on Denman Island, BC, Canada. Central Park is located in approximately the middle of the Island. Central Park is ~147 acres in size and contains two large wetlands and a recovering forest after years of logging in the area. It was logged by horses in 1998 and then heavily logged in the year 2000. The forest is specifically a Coastal Douglas-fir. This kind of forest is relatively rare in B.C. and is threatened. Local conservationists have identified up to 64 different birds that have been spotted in this park.

I visited this area first on June 24th, 2019 between 6:28 and 7:37 PM. On this day the weather was observed to have a low of 12 degrees C and a high of 19 degrees. It was sunny with clouds.

While on this walk I noticed 3 interesting potential study areas with the local ecology:

1.) Some but not all arbutus trees appeared to be dying or suffering from some ailment. Some trees had dark covered bark and leaves, while others had very little or no sign of damage. Arbutus are known to shed leafs and bark at various times in the year, but my observations were outside the normal leaf shedding. Why were some of these arbutus trees affected but not others, and why were some of them dying?

2.) In some areas ovate shaped leaves appeared covered in small holes. What organism or local weather or climate caused these holes in these particular areas?

3.) In some areas of the forest and meadows, there appeared to be large concentrations of thistle plants (dozens in a small parcel of land). Why were there so many thistle species in such a small area?

Field Observations:

Blog Post 5

TRU Open Learning Leave a comment

User:  | Open Learning Faculty Member: 

Background: I completed five transects each five meters apart.  Each transect consisted of five 1m2 sampling areas alternating from left to right 1 meter apart. For a total of 25 sampled spots. My sampling unit was the presence of fomitopsis pinicula in a 1m2 area. I only found 4 of the fungi in my sampling area. They were all living on dead tree stumps.  I think the presence of the fungi is very dependent on the location in the forest because it mostly grows on dead stumps. From my observations certain areas of the forest have very few stumps compared to other areas.

 

Difficulties: There were several problems I ran into:

  1. I used a measuring tape to measure out the distances which was time consuming and inaccurate at measuring the 1m2
  2. My study area is on a mountain which is difficult to walk up and down making the process arduous.
  3. I did not find very many samples of the fungi which I know can lead to statistical errors because of the low number of samples.
  4. The 1m2 was used because this is the most common unit I saw. I think because the fungi are not like grasses or other plants that can grow everywhere on the ground coverage this may not be an efficient unit to measure.

 

Modifications: I plan on making the following modifications:

  • Increasing the transect size to cover more ground. I think increasing it from 5 meters to 10 or 20 meters. An increase in my unit from 1 m 2 may also help.
  • Using a pre measured string to measure the areas to have more accuracy in the measurement.

 

The modifications would hopefully help me find more of the fungi and have more accurate measurements.

Post 4 – Sampling Strategies

TRU Open Learning Leave a comment

User:  | Open Learning Faculty Member: 

For the sample strategies, I used area based sampling techniques.  The sampling time was actually very close to the same for all three strategies, with systematic being the fastest at 12 hours, 37 minutes, random being second at 12 hours, 39 minutes, and finally haphazard at 13 hours, 1 minute.  The percentage error for the two most common species found systematic and random to be the most accurate.  But looking just at eastern hemlock, the most common species, Random was the most accurate, then systematic, and finally haphazard.  The percentage errors found in the two rare species varied more so, with systematic actually being the least accurate overall.  However, looking at just the most rare species, white pine, Random was the least accurate, as there were no white pine found in those sample plots, systematic was next, and haphazard was the most accurate.  The accuracy generally decreased as species abundance decreased, which is logical, as some sampling strategies found none of the rare trees, making their percentage of error 100%, which is more likely to happen the less frequent a species is.  When the species abundance is higher, systematic and random sampling appear to be more accurate, while haphazard just happened to more accurate in rare species.  The transect lines being used (systematic) for rare species is the least accurate as the area is large, and following a single bearing is unlikely to capture all that the area encompasses.  However, if the area was stratified before sampling, a transect line would be as accurate.  So, all the sampling strategies are likely close to the same accuracy, and if done properly, all strategies likely have very similar accuracy.

Post 9: Field Research Reflections

TRU Open Learning Leave a comment

User:  | Open Learning Faculty Member: 

Upon reflection of my study, there are many things that could have been changed in order to make the study more critical with less room for error. Most of these errors could have been avoided but the concept and repercussion of the error were not realized until it had been made. Implementing the design that was originally derived was simple enough but lead to problems when it became evident there were many possible predictor variables at play. This sample of the practice of ecology has led to a significantly greater understanding when it comes to the scientific process and ecological theory. The process of turning an idea into a testable hypothesis is now much clearer. It is also evident how important the scientific process is in regards to ecology. Without the scientific process ideas around ecology would remain ideas and never make it to scientifically proven fact.

 

 

 

Blog Post 4

TRU Open Learning Leave a comment

User:  | Open Learning Faculty Member: 

 

Technique: Systematic Sampling of Area

Sampling Time: 12 hours 7 minutes

1 Most Common Species: Eastern Hemlock

Actual Density: 469.9

Data Density: 425.0

Percent Error: 9.5%

2 Most Common Species: Sweet Birch

Actual Density: 117.5

Data Density: 95.8

Percent Error:    18.4 %

 

1 Rarest Species: white pine

Actual Density:  8.4

Data Density:  0.0

Percent Error: 100 %

2 Rarest Species: Striped Maple

Actual Density:  17.5

Data Density:  37.5

Percent Error: 114 %

Technique: Random Sampling of Area

Sampling Time: 12 hours 13 minutes

1 Most Common Species: Eastern Hemlock

Actual Density: 469.9

Data Density: 369.6

Percent Error: 21%

2 Most Common Species: Sweet Birch

Actual Density: 117.5

Data Density: 82.6

Percent Error: 29%

 

1 Rarest Species: white pine

Actual Density:  8.4

Data Density:  13.0

Percent Error: 54.7

2 Rarest Species: Striped Maple

Actual Density:  17.5

Data Density:  0.0

Percent Error: 100%

 

Technique: Haphazard Sampling of Area

Sampling Time: 13hours 37 minutes

1 Most Common Species: Eastern Hemlock

Actual Density: 469.9

Data Density: 430.8

Percent Error: 8.3%

2 Most Common Species: Sweet Birch

Actual Density: 117.5

Data Density: 126.9

Percent Error: 8.0 %

 

1 Rarest Species: white pine

Actual Density:  8.4

Data Density:  0.0

Percent Error: 100%

2 Rarest Species: Striped Maple

Actual Density:  17.5

Data Density:  34.6

Percent Error: 97.7 %

 

 

Conclusion: Systematic sampling was the fastest technique at 12 hours and 7 minutes; however, Random sampling was close behind at 12 hours and 13 minutes. The accuracy was affected by species abidance the more rare species percent error was easily skewed by not finding any of the species such as white pine in a sampling regiment. With more samples the accuracy improved. All of the sampling techniques were similar in that the rare species had very high percent errors. Random sampling had slightly lower overall percent error.

Post 3 – Ongoing Field Observations

TRU Open Learning 1 Comment

User:  | Open Learning Faculty Member: 

As I stated earlier, Cottonwood park is a place I go to very regularly to walk my dogs.  I have been having troubles pinning down an organism or an attribute that I can study, as it cold, there is 10-20cm of snow on the ground, and the majority of the plant species are dormant or hidden, and identification between branches can be difficult.

One thing I noticed were the rosa sp. species, as some still have their rose hips persisting through the cold.  It may be possible that there are actually two different species of rose,  Rosa acicularis (Prickly rose), and Rosa nutkana (Nootka rose).  I thought if I could differentiate between the two, I could study if one species had more rose hips persisting through winter than the other.  However, differentiating between these closely related species is difficult even when there are leaves and flowers, let alone when there are just decaying fruiting bodies.

Moving forward, I made some more field observations this morning (December 24).  I walked the loop on the east end of my study area from 6:00-7:00am.  The temperature was -7 degrees, with a wind chill of -15 degrees, and the sky was overcast.  It was still dark and everything is always very quite at this time.  It is rare that I see anyone.  However, this is the time of day where I usually see the local fox and hares.  Unfortunately, I did not see either this morning, but I did see a high amount of hare tracks, of all different sizes (Figure 1).  I see a surprising amount when walking in the morning, as my headlamp allows for attributes to stand out.

Figure 1: Hare tracks in the snow.

Though I find the animal aspects of the park very interesting, I think it would be troublesome to study as I will always have my dogs with me when I am at Cottonwood, which usually leads to most wildlife running away.

Next, I thought I would focus on the flora of the area.  Concentrating on the island portion of the park, I noticed that there were some areas that contained more coniferous tree species than others.  This area is predominantly Cottonwood (Populus balsamifera), with large veterans in the overstorey, and the multiple clones in the understorey, as well as many other riparian type woody shrubs and herbaceous plants (Figure 2).

Figure 2: Left, photo showing large veteran Cottonwood trees; Right, photo showing more coniferous trees.

With that said, I have decided to study the density of coniferous trees on different parts of Cottonwood island.  I am going to use three areas.  There is an inlet from Nechako river that makes a portion of this park an island.  I am going to use this as my gradient.  An area at the intake of the inlet, one in the middle, and one at the end of the inlet (Figure 3).  My initial observation is that there are more coniferous trees around the middle section of the island inlet.  I believe this is because the areas near the beginning and the end of the inlet are narrowing and closer to the Nechako and inlet water, increasing soil moisture , thus increasing the density of Cottonwood trees that are able to out-compete the coniferous species.  I believe there is a relationship between the density of understorey cottonwood and the density of coniferous species, and my hypothesis is, that in areas with less cottonwood regeneration, there will be more coniferous trees.

Figure 3: Locations at Cottonwood park island along the gradient.

The response variable in this study are the Coniferous trees and the predictor variable are the number of regenerating cottonwood stems.  Both my response variable and predictor variable are categorical as they will be classified  into presence/absence counts.

Blog Post 5 : Design Reflection

TRU Open Learning Leave a comment

User:  | Open Learning Faculty Member: 

Blog Post 5: Design Reflections

It was difficult to measure different sample area, every time I observed. Each area, preserved hill, ornamental steps, ornamental gardens were not really huge, randomly selected quadrat easily overlapped the area that I already observed, which increases error in data accuracy. Therefore, I though it might have been better if it followed systematic sampling techniques rather than random sampling technique. Also, in preserved hill there was not much dry/green grass, barely any vegetation could be seen. In ornamental steps green grass did not exist however, green clover was easy to observe. Although there were a lots of points that were missing however, the results quantitatively supported my hypothesis that there were more fresh plants observed as the landscape was more intervened with artificial modification.

To make more accurate, credible and supportive data I would make several changes. First, I will change my sampling technique from random to systematic. Also, considering the environment there might not be an vegetation the denominator will be the area of the quadrat 100 cm^2. Finally, considering other species of vegetation exists, I would change the numerator will be changed to the green surface area in that quadrat. This will test more closer to the hypothesis I was about to test, also it will reduce the error, and the conditions of sampling is more refined to support the hypothesis.

 

Post 4: Sampling strategies

TRU Open Learning 1 Comment

User:  | Open Learning Faculty Member: 

 

 

Haphazard or subjective sampling took the least time while sampling because sample selection didn’t took a lot of time because it was chosen randomly compared to other sampling techniques.

Two most common species were Eastern Hemlock, Sweet Birch and two rarest species were Striped Maple, White Pine. In Systematic sampling techniques percentage error for these four species; Eastern Hemlock, Sweet Birch, Striped Maple and White Pine each were: 1.6%, 5.6%, 100%, 100%. In Random Sampling technique, 20.2%, 11.3%, 18.8%, 50%. In haphazard or subjective sampling, 10.9%, 45.2%, 54.3%, 100%. Generally, the results turned out to be more accurate if the species were abundant. Only in abundant species systematic sampling was accurate than other sampling techniques. However in overall range, random sampling technique was overall most accurate compared to two sampling techniques.

 

Blog Post 3 : Ongoing Field Observations

TRU Open Learning 1 Comment

User:  | Open Learning Faculty Member: 

The organism that I decided to mainly observe were several types of grass which were mainly; long lawn grass, clover, short zoysia grass.

Three location I chose to make observation was; Preserved hill nearby the church, Ornamental stone steps in apartment complex, Ornamental garden in the apartment complex. Those three areas had been selected based on how the gradient of ornamental construction process. As it goes from preserved hill to ornamental garden land development gradient gets denser.

Observation of the area was based on three categories; how equally types of grass were distributed per area, how abundant each species were, and finally each characters of the species depending on the area. First in conserved hill, distribution of grasses were not even and species were barely observed, all I could observe was big rocks and dry trees and fell down leaves. Barely seen grasses were dry already, it just shown certain trees and that was all. Secondly in Ornamental stone steps, grasses were fairly distributed evenly but it was majorly covered with green healthy clovers rather than other types of grass. Although some short grass was observed they were dry already. Finally, in Ornamental garden the area was majorly covered with long lawn grass, they were all green and healthy, they were also distributed evenly along the trees and bushes and the species were majorly long lawn grass. It was hard to believe that these plants are growing in this cold whether.

After the observation, I assumed that as more artificial work was observed, the species were more abundant and fresh, compared to natural region. It might be because of the intention of it, ornamental function of the area. Due to this thought, I could come up with and hypothesis that, as more human induced changes occurs in land, the types of species are more abundant, and they grow more fresh compared to natural regions.

Potential responser variable that I could measure would be types of species observed per area and this variable is categorical. Potential explanatory variable is region that has got more range of intervention of land changes, this variable is continuous. And in this case potentially, logistic regression would be an ideal experimental design.

Blog Post 8: Tables and Graphs

TRU Open Learning Leave a comment

User:  | Open Learning Faculty Member: 

I did not have too much trouble organizing my data. I estimated percent cover of English ivy in each of my quadrants and used the percent cover table to record these percentages as various cover classes and used the midpoints of each of the classes to compute the mean of all the replicates in each of the substrates. I then used a bar graph to represent the mean cover for each substrate. This data looks as I expected it to and I did not see anything surprising. I performed a one way ANOVA test between the two substrates and found a significant difference.