Post 3: Ongoing Field Observations

User:  | Open Learning Faculty Member: 


1) I plan to investigate the presence of green trunk pigmentation in aspen trees. I would like to determine if it is present on all aspen trees or if there are factors affecting the distribution.

2) My initial observation area only contained one small stand of aspen trees. Since aspen trees in close proximity to one another are often clones of each other, aspen trees from other areas would need to be observed in order to have independent replicates. I chose five different areas to observe aspen trees. The distance between the areas ensures the populations were independent of each other. Once I arrived at an area, the trees were chosen haphazardly mainly due to accessibility. I also had to choose areas that were accessible at this time of year by car as well as areas that had trees in snow shallow enough to be accessed on foot. Since accessibility was a concern, all the sample areas were located in close proximity to roadways or plowed parking lots.

In my first sample area I observed that the green color was present on the south east side of the trees but was not visible on the remainder of the trunk circumference. I observed three trees in this location and found them all to have the same green distribution pattern. The trees in the first area were in an open field on the edge of a cliff above a riparian area and had no competition for light or nutrients. Based on the observations from the first area, I wanted to see if different levels of competition/vegetation type, elevation, or direction had any impact on the presence of the green color.

The second area I observed was approximately 1.5 km from the first. The vegetation in this sample area was populated with both coniferous and deciduous trees. This type of vegetation provided much more competition for light and nutrients than was present in the first area. Five trees were observed at this location and all had green pigment present. The coloration was observed to be mainly on the south east sides of the trees and didn’t start until approximately three feet up the tree. Two of the trees were in denser areas of the forest and the green color was observed to be all the way around the trunks on these two trees. The bark on the larger, older trees was not green on the north side since this side is typically very rough and black so green pigment would not be visible.

The third area was the furthest away and the highest elevation. The trees in this area were again in a forested area with competition as well as being in a harsher environment due to the elevation. Winter in this area typically lasts longer and has much more snow than the other areas sampled. All the trees sampled in this area were green all the way around the trunk. One tree on the edge of a clearing had darker green coloring on the north side of the trunk. This tree was on the edge of a small clearing and had no immediate competition on the north side.

The fourth area was similar in climate to the third area but the trees in the fourth area were in the open with minimal competition. The trees with the smaller circumference in this area were found to be green all around and for most of the height that was observable with the south east side being a darker shade of green. The larger trees were a much lighter shade of green. One of the trees sampled in this area was next to a light that came on at dusk and off at dawn. This tree had very limited green visible.

The fifth area was on the edge of a marsh. The trees in this area were in the open with no tree competition and plenty of access to resources. They were located in town so may also have be impacted by artificial light. The green on these trees was noticeably darker on the south west side of the tree. The trees in this area did not have black scabby bark on the north side and they had branches growing almost uniformly on all sides. The trees in this area were observed to have green pigment present on their branches as well.

3) The patterns I observed could be caused by competition from surrounding vegetation, access to sunlight, exposure to artificial light, age of tree, or elevation.

My hypothesis is that green pigment will be present on the south side of all aspen trunks. I predict that the aspen trees will have green pigment on the south side of the trunk to maximize its access to sunlight.

4) One response variable is the presence of green coloring on the trunk of the aspen tree. An explanatory variable would be surrounding competition for sunlight. In this experiment the response variable would be categorical and the explanatory variable would be continuous.

One thought to “Post 3: Ongoing Field Observations”

  1. Your hypothesis and prediction are very clearly stated. Considering the age of the trees is one of the factors that may cause the observed pattern, will you be looking at both mature and young trees? In the fourth area, you describe when the sun hits the tree, will you be taking this too into account for your different tree samples?

    Instructor: Robyn Reudink

Leave a Reply to TRU Open Learning Cancel reply

Your email address will not be published. Required fields are marked *