Results & Discussion
1. Does fire severity affect the regeneration of aspen stands?
From the exploratory data we can see that moderate severity fires tend to promote the most aspen root suckering response, however, the results are not statistically significant. Statistically, this proves my hypothesis wrong. Differing levels of fire severity do not effect aspen trees enough to cause significant differences in regeneration. However, this was a very small study and this could be studied further. |
2. Does habitat type (i.e. bog or upland) have an effect on regeneration of black spruce stands?
Original data exploration shows that habitat type does not have a significant effect on the number of black spruce seedlings regenerating. Bogs tended to have higher numbers of seedlings, which was very surprising to me as I predicted bogs to have fewer ideal microsites for black spruce seedling establishment. However, upland black spruce sites facilitate a significantly higher number of aspen regeneration. This could be the reason why black spruce seedlings didn't do as well on the upland sites as I expected, as they could have been out-competed by the aspen. To further understand the distances between the groups of habitats and fire severities, a perMANOVA was performed. This showed that burnt upland black spruce had similar tree regeneration to an aspen stand that burnt at low severity (Table 1). Figure 1 also shows these results, with many upland black spruce sites looking more similar to aspen stands that black spruce stands. We can also see that bog black spruce stands are more similar to the control black spruce stands, and that aspen stands that experienced low severity burns are more similar to the control aspen stands.
This is a very interesting result to me, as I was not expecting aspen to be growing on black spruce sites as much as they were. This will need to be studied further and these sites should be returned to in a few years. It would be very beneficial to determine if the aspen will persist, and the forest type will change, or if this presence of aspen is just a stepping stone in the successional trajectory of a black spruce stand here in Alberta. Black spruce are an understudied species in Alberta, and this knowledge will valuable for future forest management.
Original data exploration shows that habitat type does not have a significant effect on the number of black spruce seedlings regenerating. Bogs tended to have higher numbers of seedlings, which was very surprising to me as I predicted bogs to have fewer ideal microsites for black spruce seedling establishment. However, upland black spruce sites facilitate a significantly higher number of aspen regeneration. This could be the reason why black spruce seedlings didn't do as well on the upland sites as I expected, as they could have been out-competed by the aspen. To further understand the distances between the groups of habitats and fire severities, a perMANOVA was performed. This showed that burnt upland black spruce had similar tree regeneration to an aspen stand that burnt at low severity (Table 1). Figure 1 also shows these results, with many upland black spruce sites looking more similar to aspen stands that black spruce stands. We can also see that bog black spruce stands are more similar to the control black spruce stands, and that aspen stands that experienced low severity burns are more similar to the control aspen stands.
This is a very interesting result to me, as I was not expecting aspen to be growing on black spruce sites as much as they were. This will need to be studied further and these sites should be returned to in a few years. It would be very beneficial to determine if the aspen will persist, and the forest type will change, or if this presence of aspen is just a stepping stone in the successional trajectory of a black spruce stand here in Alberta. Black spruce are an understudied species in Alberta, and this knowledge will valuable for future forest management.
Table 1. PerMANOVA pairwise comparison, with Holm's adjustment, for tree regeneration density.
Figure 1. PCoA, with Bray-Curtis distance calculation, ordination showing differences in tree regeneration across fire severities, habitat types, and control stands.
3. Do stand characteristics (i.e. age, height, basal area, competition) have an effect on regeneration in aspen and black spruce stands?
After exploring stand characteristic interactions with stand regeneration using basic ANOVAs, scatter plots, and correlation analyses it was found that stand age, canopy height, and basal area did not have any significant effect on regeneration, while competition from understory plant species did have significant effects.
To explore this further a regression tree analysis was performed for each stand type (aspen and black spruce) using all the stand characteristic data. Figure 2 shows that in aspen stands, forb cover accounts for the most variance explained (40%) follwed by grass cover and stand age. Figure 3. shows that in black spruce stands, bryophyte cover accounts for the most variance explained (39%) followed by lower levels of bryophyte cover and forb cover. It is interesting to see that black spruce tend to thrive with higher amounts of competition, while aspen tend to be less productive with higher amounts of competition. This, in part, is due to the silvics of each tree species. Black spruce are a shade-tolerant species and are able to do well without high levels of light, while aspen are a shade-intolerant species that experiences high levels of mortality when it does not have access to enough light. Higher amounts of cover in the black spruce stands, particularly bryophytes, shows that the habitat is more suited for black spruce rather than aspen because these sites with higher amounts of bryophyte cover are characterized as being wet which is not ideal for aspen growth.
The results from the regression tree analyses match the results from the data exploration. Understory vegetation cover has the greatest effect on stand regeneration, while other stand characteristics do not. This is likely do to the fact that stands were chosen to be similar to one another to get better comparisons on the effects of fire severity and habitat types. Age did come out to explain some variance in the aspen stands, with younger stands having greater amounts of regeneration. This could be due to heart rot, which is a common disease for older aspen trees to develop, and the unhealthy trees may not be able to regenerate as well as healthy trees.
After exploring stand characteristic interactions with stand regeneration using basic ANOVAs, scatter plots, and correlation analyses it was found that stand age, canopy height, and basal area did not have any significant effect on regeneration, while competition from understory plant species did have significant effects.
To explore this further a regression tree analysis was performed for each stand type (aspen and black spruce) using all the stand characteristic data. Figure 2 shows that in aspen stands, forb cover accounts for the most variance explained (40%) follwed by grass cover and stand age. Figure 3. shows that in black spruce stands, bryophyte cover accounts for the most variance explained (39%) followed by lower levels of bryophyte cover and forb cover. It is interesting to see that black spruce tend to thrive with higher amounts of competition, while aspen tend to be less productive with higher amounts of competition. This, in part, is due to the silvics of each tree species. Black spruce are a shade-tolerant species and are able to do well without high levels of light, while aspen are a shade-intolerant species that experiences high levels of mortality when it does not have access to enough light. Higher amounts of cover in the black spruce stands, particularly bryophytes, shows that the habitat is more suited for black spruce rather than aspen because these sites with higher amounts of bryophyte cover are characterized as being wet which is not ideal for aspen growth.
The results from the regression tree analyses match the results from the data exploration. Understory vegetation cover has the greatest effect on stand regeneration, while other stand characteristics do not. This is likely do to the fact that stands were chosen to be similar to one another to get better comparisons on the effects of fire severity and habitat types. Age did come out to explain some variance in the aspen stands, with younger stands having greater amounts of regeneration. This could be due to heart rot, which is a common disease for older aspen trees to develop, and the unhealthy trees may not be able to regenerate as well as healthy trees.
Figure 2. Regression tree analysis for tree regeneration in aspen stands, using all stand characteristic variables.
Figure 3. Regression tree analysis for tree regeneration in black spruce stands, using all stand characteristic variables.
4. Does fire severity and habitat type have an effect on the regeneration of ground cover vegetation?
To further explore how fire effects ground cover vegetation a perMANOVA and MetaMDS ordination were performed to understand the differences in distance between groups. Table 2 highlights interesting results from these tests, these results can also be visualized in figure 4. Firstly, we can see that the understory is significantly different in moderate and high severity burns compared to the control stands, while low severity burns had similar ground cover to the control aspen stands. This was expected, as low severity burns burn very little of the forest floor. We can also see that both burnt black spruce habitats has significantly different ground cover compared to the control black spruce stands. Very interestingly, burnt bog black spruce is actually very similar to the control aspen stand. These results show that fire homogenizes understory plant communities quite well. It is also important to note the large spread of control bog black spruce stands. This shows that they can very greatly in composition, which could have an effect on our results.
To further explore how fire effects ground cover vegetation a perMANOVA and MetaMDS ordination were performed to understand the differences in distance between groups. Table 2 highlights interesting results from these tests, these results can also be visualized in figure 4. Firstly, we can see that the understory is significantly different in moderate and high severity burns compared to the control stands, while low severity burns had similar ground cover to the control aspen stands. This was expected, as low severity burns burn very little of the forest floor. We can also see that both burnt black spruce habitats has significantly different ground cover compared to the control black spruce stands. Very interestingly, burnt bog black spruce is actually very similar to the control aspen stand. These results show that fire homogenizes understory plant communities quite well. It is also important to note the large spread of control bog black spruce stands. This shows that they can very greatly in composition, which could have an effect on our results.
Figure 4. NMDS, with Mahalanobis distance calculation, ordination showing differences in understory vegetation cover across fire severities, habitat types, and control stands.
Table 2. PerMANOVA pairwise comparison, with Holm's adjustment, for understory vegetation cover.