Do alpine hiking trails show
evidence of edge effects?
evidence of edge effects?
Data & Preliminary Analyses
Data
Figure 7: Sample data table
Data table:
Figure 7 is a simplified version of my dataset. Treatment is the predictor variable, while slope, aspect and soil compaction are covariables. Species and non-vascular/abiotic cover types are response variables. Only a small portion of the overall species list is included in this simplified table. Transect represents a random variable (block).
Data analysis:
Analyses for this project will consist of two parts: preliminary, exploratory analyses, and formal multivariate and univariate analyses. The objective of the preliminary exploration is to visualise trends in the data and provide direction for the formal analyses.
Figure 7 is a simplified version of my dataset. Treatment is the predictor variable, while slope, aspect and soil compaction are covariables. Species and non-vascular/abiotic cover types are response variables. Only a small portion of the overall species list is included in this simplified table. Transect represents a random variable (block).
Data analysis:
Analyses for this project will consist of two parts: preliminary, exploratory analyses, and formal multivariate and univariate analyses. The objective of the preliminary exploration is to visualise trends in the data and provide direction for the formal analyses.
Preliminary Analyses
Soil compaction:
Although bulk density of soil varies naturally, soil compaction can give an idea of how heavily an area is trampled (Tejedo et. al. 2009; Cole 1987). Though it cannot give quantitative measures of trampling pressure (i.e.: number of passes), higher soil compaction indicates that an area is has likely experienced more trampling than an area with lower soil compaction.
Although bulk density of soil varies naturally, soil compaction can give an idea of how heavily an area is trampled (Tejedo et. al. 2009; Cole 1987). Though it cannot give quantitative measures of trampling pressure (i.e.: number of passes), higher soil compaction indicates that an area is has likely experienced more trampling than an area with lower soil compaction.
Figure 8: Soil compaction
Figure 8 shows that soil compaction declines along the distance gradient from the trail, suggesting that trampling pressure is highest near the trail and declines gradually with increasing distance.
Individual species responses:
In a previous analysis I found that total vascular cover does not change with increasing distance from the trail. Therefore, if the vascular community is impacted by trampling, it must be in the form of relative composition rather than total cover. To begin looking into whether this is the case I have run an NMDS ordination to see if plots at different distances separate out from each other.
In a previous analysis I found that total vascular cover does not change with increasing distance from the trail. Therefore, if the vascular community is impacted by trampling, it must be in the form of relative composition rather than total cover. To begin looking into whether this is the case I have run an NMDS ordination to see if plots at different distances separate out from each other.
Figure 9: NMDS ordination
Although the majority of plots clump together, it seems that edge plots are are aggregating on the left-hand side of the graph, while the 50m plots are clumped more to the right. It is possible that this clumping is driven by a number of outlier plots, and it may be useful to identify and remove these plots to see if they are obscuring patterns in the data.
To begin looking more closely at the effects of trampling on the vascular community on the Divide, I have graphed out the the five most commonly-found species against distance from the trail to see if any of them show a progressive increase or decrease as trampling pressure diminishes.
To begin looking more closely at the effects of trampling on the vascular community on the Divide, I have graphed out the the five most commonly-found species against distance from the trail to see if any of them show a progressive increase or decrease as trampling pressure diminishes.
Figure 10: cover values for five common species
None of these species show a very clear increase or decrease associated with distance from the trail.
Non-vascular and abiotic covers:
Like vascular plant cover, total non-vascular and abiotic cover remains remarkably consistent with increasing distance from the trail. However, in this case there are clear compositional changes along the distance gradient. In this analysis I graphed out the various cover types individually. There were six non-vascular and abiotic covers in our surveys, but I will only present four of them here, because two of the cover types (moss and soil crust) did not show interesting trends. The remaining cover types - rock, soil, lichens and litter - will be presented in two separate graphs because the trends of the less abundant cover types are obscured when they are graphed together with the more abundant cover types.
Like vascular plant cover, total non-vascular and abiotic cover remains remarkably consistent with increasing distance from the trail. However, in this case there are clear compositional changes along the distance gradient. In this analysis I graphed out the various cover types individually. There were six non-vascular and abiotic covers in our surveys, but I will only present four of them here, because two of the cover types (moss and soil crust) did not show interesting trends. The remaining cover types - rock, soil, lichens and litter - will be presented in two separate graphs because the trends of the less abundant cover types are obscured when they are graphed together with the more abundant cover types.
Figure 11: Non-vascular cover trends
We see an increase in litter and lichens and a decrease in rock and soil we move further away from the trail.
