Category: Post 8: Tables and Graphs

Blog Post 8: Tables and Graphs

TRU Open Learning Leave a comment

As my data collection I found 5 replicates of Arbutus trees by splitting 5 groups of clustered trees in different geographic regions, and then I assessed the health of each tree by estimating the proportion of healthy leaves on the tree.

I then took this data and graphed it producing a graph relating the % of healthy leaves vs each of the five groups to see if I could notice a trend. I then averaged each replicate (2 trees) per group to find the average of the leaf health in each group.

My hypothesis was that the poor health of the Arbutus trees along this route was primarily caused by poor access to sunlight and resources in the soil directly caused by competition with taller and faster growing neighbouring trees.

My prediction was that the regions that had more open areas, a shorter tree canopy, or niches where only Arbutus trees could easily grow (cliffs), that these regions would have the largest trees and the healthiest leaves.

This pattern was observed in the graphed data. As the landscape transitioned from a dense forest to open fields to elevated cliffs, the leaf health of the Arbutus trees did increase. The most open areas and those with cliffs had the greatest leaf health, those with the thickest forest had the worst health, and a zone in-between the two had moderate levels of leaf health, which is also to be expected.

Blog Post 8: Tables and Graphs

TRU Open Learning Leave a comment

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.

Post 8: Tables and Graphs

TRU Open Learning Leave a comment

For my project regarding canopy closure and deciduous trees, I chose to do a bar graph. At first, I had a hard time deciding which type of graph showed the data effectively without confusion. This was a bit of a challenge due to having a large amount of data. In the end, I chose a bar graph as it clearly showed all the necessities while still providing a clear understanding of what was happening. The outcome of my data was mostly as expected, the only thing that surprised me was how little deciduous trees resided in a canopy closer <60%. I knew that the numbers would be significantly less than in open canopy areas but the fact that none were present <30% was truly surprising. if the study were to be conducted again more accurate canopy closure should be measured using photography. Due to randomized plot location, our data set was somewhat narrow with significant data lacking in canopy closure of 40-60% range. In future studies, non-randomized plots could also be implemented to ensure a broad data set was derived.

Post 8: Tables and Graphs

TRU Open Learning Leave a comment

Prior to going into the park, I already knew that I wanted to collect my data in table format versus writing a summary for each quadrat. I did not have initially know what data I wanted to collect until I went to an area of the field where some dandelions were located. This initial assessment was not included in my final data collection but served as a way for me to decide what information I wanted to take note of when summarizing each quadrat.

I eventually decided on the following columns for my table: location of quadrat, total number of flowers, flower type, number of each flower type, and other notes.

My results were as expected because the northern (both northeastern and northwestern) areas of the park had the greatest number of flowers. These results were similar to my initial observations when I only included dandelions. Surprisingly, the centre quadrat had more flowers than I expected, considering it is a flat-lying area and is typically walked over often by both humans and animals.

Blog Post 8: Tables and Graphs

TRU Open Learning Leave a comment

I chose to use a table to represent the field data I collected in this study. My table depicted a summary of each species I found at the range of elevations along the slope, and the percent canopy coverage of those species. Also included was quadrat size, location, and area covered by each species. Many species appeared in more than one elevation location. Lower elevations were dominated by pine grass and clover. Mid elevations primarily displayed the common fern and Saskatoon berry bushes. Finally, the highest elevations I recorded data for (10-11 m above the base of the slope) were dominated by the Lodgepole Pine and Paper Birch. I had no difficulty in organizing and aggregating the data. I may be able to summarize the information more concisely in graph format for the final report. In graph form, I would be able to show how individual species percent canopy cover changes across the entire slope in a more understandable visual way, rather than listing the species found at each site and their coverage. The outcomes of this data conformed to my expectations, nothing new was revealed. As predicted, more complex plants were more abundant at higher elevations, perhaps due to more sunlight exposure, or a change in nutrients found upslope. For further exploration, a more comprehensive study could be undertaken to determine if this pattern applies to the entirety of Terrace Mountain, or only on this specific slope.

Blog Post #8 – Tables and Graphs

TRU Open Learning Leave a comment

I initially made a table noting down the percentage of leaf color change for each of the five sections for the three trees I’m observing. I realized that the table was going to very long and almost a bit unnecessary. I finished it anyways and found a different way I could represent the same information. I mafe a new table with the same dates I took the observations, keeping humidity levels and time of observations. The only difference I made was making an average percentage of color change instead of having five different percentages for each tree. Now my table is more compact and easier to read.

Next I made a graph showing the color change over time. Three colors, one for each tree. It took a while to figure out how I was going to graph this and a way to assess it. There may be some changes later, bit everything seems pretty clear now.

The outcome was similar to what I expected, but I did wonder why these trees changed color slower than all the other trees. Could it have something to do with being in a highly maintained park by the city or is it because they were bigger trees compared to the other and were more able to collect resources they needed? Maybe it could be another reason too.

Blog Post 8: Tables and Graphs

TRU Open Learning Leave a comment

The graph I submitted for my Small Assignment 5 illustrates the relationship between slope incline (%) and common snowberry density (stems/m2). I added a linear trend line to my line graph to visually show that as slope incline (%) increases, the density (stems/m2) of common snowberry decreases. To produce this graph, I stratified my slope incline (%) into four distinct ranges. The ranges were 0-5%, 6-10%, 11-15% and 15%+. I determined these ranges based on my data collection. I then manipulated my data by changing the number of stems per quadrat I collected into density (stems/m2) by dividing my stems per quadrat by 2.25m2. I then calculated the average density in each slope incline (%) range, for example between 0-5% slope incline, the average density of snowberry was 15 stems/m2. Between 6-10% slope incline, the average density of snowberry was 8 stem/m2.

The linear trend line on my graph illustrates the general trend I was predicting in support of my hypothesis, that snowberry distribution is determined by slope incline (%). Specifically, I predicted that snowberry will be present in area where slope is less than 20% and that snowberry density will decrease as slope incline (%) increases.

When I was first organising my data and producing graphs, I didn’t think my results were showing the trend I predicted, and my graphs appeared cluttered with too much information. As I started to aggregate my data into different ranges and averages, my graphs appeared to show a better trend and I think they are easier for the reader to interpret.

As I am working through my data, I am noticing some trends that I didn’t predict, for example my data is showing that common snowberry is highest in Site 1 Eastern Area compared to Site 2 Riparian Area. During my initial field observations, I expected common snowberry to be at highest density in the riparian area. My data is also showing that light exposure is similar in Site 1 and Site 2 compared to Site 3 Upland Area, which could be another variable determining snowberry distribution. My soil moisture data did not show what I expected, where Site 3 Upland Area was not the driest site, where I was expecting Site 1 and Site 2 to have the highest soil moisture, and Site 3 to have the lowest, however this is not the case with my data. I also want to evaluate slope aspect (degrees) as a predictor variable.

As I am working through my final report, I will be outputting more graphs that will hopefully further support my hypothesis and show other potential trends.

Blog Post 8: Tables and Graphs from Whispering Woods

TRU Open Learning Leave a comment

I decided to create a graph illustrating the differences in mean soil moisture among P. tremuloides trees located at the bottom (n=10) and top (n=10) of Whispering Woods hill across the four times I collected data. Initially I had difficulties visualizing this graph because it required two lines on one graph: one for the means from the top of the hill trees, and one for the means from the bottom of the hill trees. I also had difficulties with the y-axis label because the soil moisture probe I used to measure soil moisture did not specify its units, thus the best I could do was treat it as a “relative” soil moisture level where 10.0 was the wettest and 0.0 was the driest. This is easy enough, but made determining the “units” for the y-axis more difficult. I decided to explain my choice in “units” in the figure caption.

The outcome was expected, as across all four data collection sessions the mean soil moisture around the trees at the bottom of the hill were higher than at the top of the hill. The data also revealed that the relative difference in mean moisture is similar, regardless of what the actual moisture ratings are. For instance, on my third data collection session, I visited Whispering Woods earlier in the morning than on my other dates, so all of the moisture readings were lower than usual due to the cold temperature. Regardless of this, the relative difference in mean soil moisture was still similar to data collected later in the day, sitting at a difference of about 2-3 soil moisture levels. This is an interesting finding that I will further explore in my literature review and final report.

That’s all!

Madeleine

Blog8 Table and Graphs

TRU Open Learning Leave a comment

Blog 8; Table and Graphs

 

October 12, 2019

 

For my field research study I created a table and a figure from the data I collected. When I collected the data I used two categorical variables. The first was the presence\absence of Hydrocotlye heteromeria in each of the quadrats I examined. The second categorical variable I used was wet\dry plot. I determined how many of the quadrats had the presence or absence of the species in each of the wet and dry plots. Since I am using categorical variables, and my study is a snapshot, natural experiment with 2 categorical variables I used a tabular method to determine my data.

 

I created an “Observed” data table including totals what the actual data I examined. I also developed an “Expected” data table and totals. I needed to create an expected data table to determine what I should have expected if the variables were unrelated. The “Expected values are what would be expected if the Null hypothesis were true and the variables were independent of one another. I then performed a Chi statistical test to determine if the relationship between Hydrocotyle and soil moisture levels was statistically significant. I determined that there is evidence that Hydrocotyle heteromeria is limited by soil moisture levels.

 

I also used the percentages of Hydrocotyle presence and absence that I found from each of the plots to create a figure. The figure is an easily identifiable way to quickly check how the counts and variables in the data table relate to one another.

 

The only problem I encountered when doing the table and the graphs was the fact that the course actually didn’t contain any material on how to process data information depending on which of the study methods we chose to use. I had to consult the professor of the course to determine the correct statistical test needed to analyse my study data. The table information and statistical testing was fairly straight forward after I completed the data counts. The outcome of the research is exactly what I had expected after reading literature. Hydrocotyle is limited by soil moisture levels. I didn’t take into account the nutrient levels in this study, but I would however like to learn more about their effect on this interesting species.