Regenerating woody species abundance varied both over time and by whether there was deer fencing protection, with initial results indicating higher abundance of oak regeneration inside the deer exclosure. Red oak density (mean trees per acre – TPA), was tested using the non-parametric Kruskal Wallis chi-squared test. Differences between TPA inside and outside the exclosures were significantly different for 2010 (X=2.66, p=0.014) and nearly significant for 2018 (X=3.04, p=0.08). Maple species abundance increased inside the exclosures and decreased outside the exclosures. Aspen abundance remained steady over time inside the exclosures while increasing slightly outside the exclosures. Birch density fluctuated with a surge in growth inside the exclosures in 2010 but dropped back down below levels outside the exclosures by 2018. Dominant shrubs inside the exclosure were honeysuckle and willow, which stayed relatively short. Dominant shrub species outside the exclosure were hazel and willow.
Inside the exclosures, understory and midstory tree regeneration in 2004 was composed of a mixture of species such as aspen, maple (both sugar and red) and birch (see Supplemental Table 4). Relative abundances of tree species varied over time, with a general decline in dominance of early seral species such as aspen. By 2005, the relative dominance of red oak was slightly more than aspen in the lower height classes (<6ft.). In 2010, maple and birch species continued to increase in height classes 4 and 5 (4-5ft.). There was continual recruitment of red and sugar maple in the lowest height classes. Red oak seedlings already in the tallest height class persisted throughout the time of the study. Recruitment of red oak seedlings into the lowest height class was constant for the entire study although they did not move into the two middle height classes.
The exclosure fencing was removed in 2013. While the red oak density declined after the fencing was removed, red maple density increased and sugar maple density remained at similar levels. Shrub dominance before 2010 was by hazel and raspberry. In 2010 bush honeysuckle became a more prominent shrub species. It is difficult to follow the changes in raspberry abundance over time as it was tallied primarily in 2004 while for the latter years only a cover class was assigned.
Outside the Deer Exclosure
Outside the exclosures, aspen was a persistent component in all height classes, but began to decline in abundance in 2010. In contrast, birch had low abundance in 2004 and 2005 but gained in abundance from 2010-2018. Maple species continuously recruited throughout the sampling period. Red oak declined slightly over the sampling period relative to the other tree species. The shrub composition outside the exclosures from 2004 – 2005 was dominated by hazel and raspberry. By 2010, the shrub layer had a more even distribution of shrub species. However shrub abundance rebounded again in 2018, but with hazel more equally mixed with raspberry and bush honeysuckle.
Plant Species Composition Trends from Relevé Plots
Diversity:
Higher numbers of species occurred on plots outside of the deer exclosures, as ruderal and competitive species were more prevalent. There were no measures of herbivory for the reference relevé plot.
Ground Flora Diversity Indices Outside and Inside Exclosure
|
Year
|
Treatment
|
S
|
H’
|
E
|
D
|
|
1994
|
CONTROL
|
28
|
2.874
|
0.862
|
0.9205
|
|
2004
|
OUT
|
38
|
3.229
|
0.888
|
0.9477
|
|
2004
|
IN
|
33
|
2.8
|
0.801
|
0.8855
|
|
2010
|
OUT
|
37
|
2.593
|
0.718
|
0.7971
|
|
2010
|
IN
|
22
|
2.48
|
0.803
|
0.8622
|
|
2018
|
OUT
|
53
|
3.581
|
0.902
|
0.9576
|
|
2018
|
IN
|
32
|
2.767
|
0.798
|
0.8857
|
Richness and dominance (S=species richness, H’=Shannon Wiener Diversity Index, E=evenness, & D=Simpson’s dominance index)
Successional Growth Changes
Vegetation on differing trajectories
The ordination analysis of the plant composition data lines relevé plots up based on how dissimilar they are from one another. The analysis showed that the ground flora was significantly different as a result of deer exclosure protection. Additionally there was a difference in how plant composition changed over time, with plots protected from deer browsing more closely resembling the 'reference’ MHc26 vegetation yet moving to a different composition. Plant composition outside of the deer exclosures contained a higher proportion of ruderal grasses and aggressive clonal herbs, including species such as Solidago canadensis, Taraxacum officinale, Vicia americana and Bromus ciliatus (see supplemental Table 6.). When looking at the figures of the ordination analysis, correlations of species data with the ordination axes (Kendall’s tau > 0.4) help identify which plants are contributing to patterns seen due to deer exclosure protection and over time. While there are some early growth stage tree species such as aspen persisting, the positive correlations for axis one and overstory trees such as sugar maple (Acer saccharum, Tau=0.6) and ironwood (Ostrya virginana, tau=0.90), as well as red and bur oak (Quercus rubra, tau=0.7 and Quercus macrocarpa, tau=0.5) indicate that there is a transition beginning to happen to a middle growth stage of MHc26. Understory species that are positively correlated to axis one are those typical of mid – late growth stages of MHc26 that tend to persist and compete well in small to medium sized gaps or closed canopies (see Table 5.) Ground flora species that help distinguish plots that were outside the deer exclosure from plots inside the exclosure were more ruderal species as mentioned earlier (see supplemental Tables 7-8). As plant composition in the plots outside the exclosure are less aligned with the 'reference' vegetation of MHc26 it is unclear if they will eventually converge and recover to a less disturbed condition in the future. Plots that are protected from deer herbivory appear more similar to the reference vegetation MHc26 yet are on a unique trajectory.
Regeneration Strategies/ Functional Groups
A secondary matrix was constructed within the analysis to explore the vegetation response to deer browse protection, year sampled, and plant functional categories based on autecological traits for successional status, competitiveness and regenerative ability. We used CSR categories defined as: C – competes in place - species adapted to low microclimatic stress and low levels of disturbance, S - stress-tolerating species - adapted to stress in climatic conditions and low levels of disturbance, and R - ruderal species - adapted to more constancy in microclimate but high levels of disturbance. Placing the management practices implemented in this case study within the context of typical disturbance regimes and considering how plants respond to disturbance and compete for resources are useful in designing future manipulation of the sites to maintain biodiversity (Bell et al. 2011). The historical disturbance regime for MHc26 included moderate surface fires, herbivory, and canopy mortality from wind events, insects, and disease that created small to large gaps (MNDNR 2006). Functional traits were assigned based on a review of plant trait databases (USDA-NRCS 2019; Kattge et al. 2011), and relevant literature (Burns and Honkala 1990; Munson et al. 2016; Novakovskiy et al. 2016b; Grime and Mackey 2002). For example, light tolerance, regenerative strategy/dispersal, and habitat preference allow for a rough assignment of categories that describe strategies plants have evolved in response to combinations of stress and disturbance intensity. The CSR categories attempt to integrate both adaptive traits for disturbance (e.g., morphology of roots and shoots, leaf forms, litter requirements, growth rates) and stress created by the variability in climatic edaphic requirements for a plant (Weaver and Grime 2007; Novakovskiy et al. 2016b).
Changes in species composition are in part caused by how plants partition available resources and how they get to and occupy optimal niches, defined by characteristics such as light, moisture, and soil (Runkle 2013; Runkle 1982). Here we use ‘regeneration niche’ categories as they relate to gap size which can be a proxy for amount of light and the availability of an open ground. Small-gap strategists are species that can fill in single-tree or few-tree gaps in the canopy through advance regeneration. Small-gap/large gap strategists are species that may either have a presence adjacent to the gap and/or fill canopy holes through colonization, or can persist with some competition with other species but require more light to attain dominance. Large-gap strategists are species that fill large canopy openings where the previous canopy no longer provides significant cover. These species are aggressive colonizers and generally require disturbed ground and full light.
Functional trait categories used in overlay matrix.
|
Response to disturbance
|
Seral Stage
|
Regeneration Niche
|
Regeneration Strategy
|
|
|
C - Comp - holds its space if already present
|
Late
|
small gaps
|
rhizomes, caudex
|
|
|
S - Stress - Can compete with occupied space and somewhat good at colonization
|
Mid-Late
|
small-gap/large gap
|
aggressive rhizomatous species and seeders
|
|
|
R - Ruderal - gets there easily but needs open space
|
Early
|
large-gaps
|
seeders, or aggressive adventitious roots
|
|
Species as indicators
The above correlations of species to the ordination axes are of interest to determine which plant species may be most useful for future monitoring efforts. Future studies would benefit from more replications of this treatment layout across similar NPCs in order to support analyses of significant plant indicators. Other studies have found plant species useful as indicators of disturbance in research on harvest impacts. For example, ruderals were found in areas with higher disturbance and continued to impact regeneration and species composition Berger et al. (2007). Other studies showed impacts to mid-tolerant species such as red oak and white pine from fern competition (Gilliam 2007, George and Bazzaz 2003).
In this case study, it appears that some species have higher competitiveness in disturbed open areas (ruderals and species that colonize paludified soils). Examples include open site colonizers such as grasses (Agrostis spp, Calamagrostis spp, and Bromus spp); species that are clonal and produce many seeds such as Eupatorium perfoliatum and Solidago canadensis; and competitors for light such as ferns like Mattuecia struthiopteris. This result is not unexpected and is similar to results from other studies where ruderal species took advantage of disturbed sites with pressure from herbivory (Wiegmann and Waller 2006). Changes in species composition to more ruderal and competitive species can be long-lasting (Gilliam 2007 and Roberts and Gilliam 2003).