Hybrid Poplar Growing in Plantations -
Growth Responses to Alternative Establishment Systems
Linkages among understory, environment and vegetation control systems
in hybrid poplar plantations
Permutational Multivariate Analysis of Variance (perMANOVA)
Consistent with trends observed using graphical data exploration with boxplots, perMANOVA showed significant differences in understory plant composition among the establishment systems (perMANOVA p = 0.001) for both years 2012 and 2013 (Table 5). All treatments were significantly different from each other (perMANOVA p = 0.001). No significant differences were found among clones (perMANOVA p = 0.519) and no interaction between treatment and clone was found (perMANOVA p = 0.962). |
Table 5. Permutational Multivariate Analysis of Variance (perMANOVA) of comparison of the understory vegetation community among establishment systems, clones and the interactions thereof for 2013.
|
Non-metric Multidimensional Scaling (NMDS)
Figure 9. Non-linear Multidimensional Scaling (NMDS) ordination for the understory composition based on functional groups (vectors).
|
Non-Metric Multidimensional Scaling with a Bray-Curtis distance measure was used to visualize the vegetation community data by functional group and to explore differences among treatments. The NMDS ordination shows the composition of the understory community grouped by functional groups (Annual Forbs, Perennial Forbs, Perennial Grasses, Cover Crops) among the establishment systems (Figure 9). In accordance with the results from the perMANOVA, the NMDS ordination shows that the vegetation, as affected by the treatments, seems to develop into distinct species communities. In addition, the NMDS shows that clones do not differ on their effect on the understory at this early stage of seedling growth. Particularly, fallow and business-as-usual systems clearly cluster cluster apart from each other and the other treatments. Notably, the clusters are at an angle towards axes 1 and 2, thus both axes are describing the variance explained in the understory community. The size of the points is related to cumulative two-year seedling diameter increment with larger points representing greater increments. |
Multivariate Regression Trees (MRT)
Multivariate regression trees were used to examine relationships between hybrid poplar seedling performance, understory community and environmental factors. Height growth, diameter growth and survival were included as response variables and understory and environmental factors used as predictor variables.
Results from the regression tree analysis indicate that clone had the largest effect on hybrid poplar performance. This is not surprising, as different clones are known to have different growth rates, resource requirements and abilities to compete for resources with the neighboring understory. Regression trees were then run separately for the two clones included in this study.
The regression tree explaining the most variation in hybrid poplar seedling performance for the Okanese clone included factors related to the environment as well as the understory plant community (Figure 10). Seedlings that were able to intercept the most light had the greatest growth increments. In contrast, seedlings transmitting most light were associated with reductions in growth performance. Furthermore, the composition of the understory affected growth performance in that a high cover of annuals was associated with higher growth increments. Annual abundance was inversely correlated with perennial abundance (data not shown). This could suggest that annuals are less competitive with trees than perennials. The last split showed that a high transmittance level of light by weeds resulted in better tree growth and suggests that competition for light was of high importance for trees.
While for the Okanese clone, a combination of environmental and understory variables as affected by the establishment systems resulted in the best MRT model, understory variables appear the most important for the Walker clone (Figure 11). Understory cover determined in great extent growth of Walker seedlings. Growth as well as survival decreased when seedlings were associated with even low cover in the first and second establishment year, including annuals and perennials, highlighting the great sensitivity of Walker to any competition with neighboring vegetation.
Results from the regression tree analysis indicate that clone had the largest effect on hybrid poplar performance. This is not surprising, as different clones are known to have different growth rates, resource requirements and abilities to compete for resources with the neighboring understory. Regression trees were then run separately for the two clones included in this study.
The regression tree explaining the most variation in hybrid poplar seedling performance for the Okanese clone included factors related to the environment as well as the understory plant community (Figure 10). Seedlings that were able to intercept the most light had the greatest growth increments. In contrast, seedlings transmitting most light were associated with reductions in growth performance. Furthermore, the composition of the understory affected growth performance in that a high cover of annuals was associated with higher growth increments. Annual abundance was inversely correlated with perennial abundance (data not shown). This could suggest that annuals are less competitive with trees than perennials. The last split showed that a high transmittance level of light by weeds resulted in better tree growth and suggests that competition for light was of high importance for trees.
While for the Okanese clone, a combination of environmental and understory variables as affected by the establishment systems resulted in the best MRT model, understory variables appear the most important for the Walker clone (Figure 11). Understory cover determined in great extent growth of Walker seedlings. Growth as well as survival decreased when seedlings were associated with even low cover in the first and second establishment year, including annuals and perennials, highlighting the great sensitivity of Walker to any competition with neighboring vegetation.
Conclusion
The results of this study into the effects of alternative establishment strategies for hybrid poplar plantations found, in accordance with literature reviewed, that the control of competing vegetation is beneficial to enhacing hybrid poplar growth rates. It should be noted that this study did not use a control that removed all competing vegetation. However, this study further demonstrated that complete eradication of weeds is not necessary, as competition with trees is dependent as much on temproal and growth form characteristscs of the competing vegetation as it is on mere presence and abundance. It was discovered that the fallow treatment was able to, for a cost similar to busisness as usual, provide good control of weeds in the first year and provide suppression of perennial weeds for the duration of the critical tree establisment period.