Research Report: Aspen Mixedwoods: complex forests are resilient forests

Mike Reinikainen, Anthony D’Amato, Shawn Fraver, John Bradford and John Almendinger (Report prepared by Mike Reinikainen and Anthony D’Amato)

Wet-Mesic Southern Boreal Mixedwoods in Minnesota

Prior to 19th century timber harvesting and contemporary management, aspen mixedwoods were older and more ‘mixed’ than they are today, with a greater diversity of tree species, size and age. Over the past century, aspen mixedwoods in Minnesota have become far more widespread, younger, less diverse and generally ‘unmixed’.

Unmixing can be attributed to both the removal of seed source through historic, selective removal of conifers and the use of short rotations to maximize aspen yields for pulpwood production.  From an aspen production standpoint, these management regimes are ideal; however, the simplification of aspen mixedwoods has negative implications for resilience (the ability of a forest to recover from a disturbance and return to its prior function). Decreases in resilience correspond to declines in water quality, wildlife habitat, and most importantly forest productivity.

Typical wet-mesic southern boreal mixedwood stand. Click for a larger version.

This study focuses on a specific aspen mixedwood type called the wet mesic southern boreal mixedwood (SBMW).  The study seeks to understand how this forest type changes in composition and structure over time and how differences in those characteristics affect productivity and stand resilience.

Contemporary SBMW dominate Northern Minnesota and are typified by aspen. Other species common in SBMW include white birch, balsam fir, white spruce, red maple and black ash. Unlike the drier, shallow soiled, fire dependent forests of the Boundary Waters region, SBMW generally occur on deeper, fine textured soils ranging from fine sandy loams to clay loams, and they generally assemble on level areas close to local water tables.

Diversity and Resilience

Why does resilience matter in SBMW?  A complex, mixed system is resilient because resources are used and production continues during or quickly after disturbance. Increased complexity of forest composition and structure could mean increased resilience for landowners whose woods include SBMW stands.

Here’s an example: When aspen is defoliated by forest tent caterpillar, balsam fir utilizes the light, water and nutrients temporarily unused by aspen. If the fir component is missing, those resources go unused, reducing overall stand productivity. Ultimately, increased complexity of forest composition and structure could mean increased resilience for SBMW landowners.

Objective and Study Design

This study was designed to better understand how simple and complex SBMW forests develop and what land managers can do to increase stand resilience and productivity. We examined tree rings in cores taken from living trees as a record of how resources fluctuate within a forest. Ultimately, such data yields tree recruitment age (when the tree began its life) and growth rate trends (a good representation of resource availability).

White spruce core showing increased growth after the death of competing trees. Click for a larger version.

We collected and analyzed 1,300 straw sized wood core samples from 10 different SBMW stands across Aitkin, Itasca, and St. Louis Counties in Minnesota. The stands ranged from 60-85 years since last harvest, and included three stand types: simple, pure aspen (A), moderate mixtures of aspen and conifers (C), and complex mixtures of aspen, conifers and other hardwoods (M).


Legacy trees: Stands with legacy trees (those that survived the last harvest, and contribute seed to the new stand) such as balsam fir, white spruce, red maple, white pine, and paper birch showed greater mixing early in stand development due to diverse advance regeneration and available seed sources at the time of disturbance. Legacy trees remain important because they continue to contribute to a diversity of advance regeneration later in stand development as short-lived trees decline and the canopy begins to break-up.

When species mixing occurs: Our results showed that the most species mixing occurred at times of increased resource availability: the initiation of the stand (year ‘0’), the end of self thinning (years ‘30-40’), and the beginning of small gap formation (years greater than ’50’). Conifer recruitment peaked approximately 30 years following harvest. Peaks correspond with the end of intense competition for resources following the self-thinning stage of young aspen.

Complexity and resilience: More complex stands can be more resilient, but a greater number of species did not always confer increased resilience. In optimum growing years when resources were abundant, more complex stands (M) produced more biomass. However, in sub-optimum years (in drought or defoliation) when resources were scarce and competition was high, less complex stands (A & C) showed slightly greater yields. This finding suggests that increased diversity does not guarantee resilience.

Instead, the type of mixture is important. Stands with mixed aspen and conifer (C) may actually be more resilient than stands lacking the conifer component because these species differ significantly in the resources they need and the areas from which they draw those resources. Competition is minimized by these differences when resources are scarce.

Management recommendations

There are a few things landowners can do to increase SBMW stand resilience and productivity.

  1. Retain legacy trees within large harvest areas or adjacent to small harvest areas.  Resilient mixtures require seed sources throughout development. Seeds from legacy trees contribute greatly to the future composition of your forest, and can maintain production in future disturbance.  If no seed trees are available, seedlings can be planted at either the beginning of a stand’s development or during periods in which gaps are likely to form due to harvesting operations or natural mortality of canopy trees (after year ‘50’).
  2. Maintain seedbed conditions required by desired species. For some species, successful recruitment depends on seedbed quality. Light seeded species like spruce will require large old dead wood for establishment or mineral soil exposure. Legacy trees and retention of old defective boles for dead wood are easy additions to a harvest plan, but mineral exposure is more difficult because the fine textured soils of SBMW require winter harvests. In addition to preventing soil compaction, winter harvests can prevent damage to advance regeneration.
Group selection harvest. Click for a larger version.

Other silvicultural systems, such as group selection harvests, may become more economical as new markets arise related to the production of energy from forest biomass. These systems are a contrast to short rotation clear-cuts, but offer an avenue to meet multiple objectives. In addition, because more complex SBMW stands offer an opportunity to increase resilience to current and future environmental stressors, these systems may allow for greater productivity in the future.

This study provides evidence for increased resilience through increased complexity and provides tools for enhancing both traits in a common, simplified aspen mixedwood. Increases in tree diversity do bolster resilience, but the types of trees present are important. Conifers are a good addition to a simple stand, and the timing of introduction should correspond to critical points of resource availability. Ultimately, establishing and maintaining a diversity of seed source is the key to long term productivity in your woods.

Mike is a graduate student in the University of Minnesota Department of Forest Resources. He can be reached at

You may also like


  1. interesting study and good ideas about an important forest type. but there seems to be little/no mention of the woody and herbaceous species that occur in these forests. they must play a role in recruitment and dynamics at most if not all stages of development–even to the point of preventing establishment of trees in some cases. thanks.

  2. Can you please provide research references to justify your assertion in the first paragraph of this article that “…aspen mixwoods were older and more “mixed” than they are today…” and “…aspen mixedwoods in Minnesota have become far more widespread, younger, less diverse…”


    Gary Erickson

  3. Gary,
    Thanks for the question. It is a broad assertion, but there are two landscape level studies that use public land survey records to better explain what pre-settlement forests looked like. Both compare pre-settlement forests to current forests as described by FIA data. This is at a large scale, but the first article points to the expansion of our aspen-birch type to nearly five times its 19th century basal area.

    Schulte, L. A., D. J. Mladenoff, et al. (2007). “Homogenization of northern US Great Lakes forests due to land use.” Landscape Ecology 22(7): 1089-1103.

    Given what we do know about the silvics of southern boreal mixedwood species and the general stages of succession in this type, I think this paper shows that mixedwoods are less ‘mixed’ and that this has to do with a landscape scale reduction in stand age (it also may have to do with a reduction of seed source, particularly white pine and white spruce). Compositionally, younger mixedwoods lean heavier towards shade intolerants like aspen and birch. Older mixedwoods should be more diverse in terms of species due to the presence of shade tolerant conifers. Ultimately, we have less of these conifer-hardwood mixtures because the stands are younger and due to a lack of seed source.

    The second study comes from some of the MN DNR’s work surrounding NPCs. Again, they use public land survey records but are making community specific comparisons to current FIA data (look at page 37 and 47).

    First, this analysis shows that older pre-settlement mixedwoods tend to be more diverse (p. 37). Second, younger age classes are currently overrepresented compared to pre-settlement forests (p. 47). It is important to note that conifer presence has been greatly diminished when compared to 19th century forests primarily due to the lack of white spruce (p.47).

    Thanks again,
    Mike Reinikainen

  4. John
    You are correct; this particular analysis gave no consideration to the recalcitrant woody shrub layer that can limit regeneration of even the most shade tolerant of our tree species. Hazel, mountain maple, and some herbaceous plants such as bracken fern can in effect halt succession to shade tolerant conifers and hardwoods when densities are high. Though not addressed, we do have shrub density information and we do know recruitment patterns of overstory trees. The next step will be relating the two.
    For discussion sake, data from this study showed peaks in tree species (balsam fir, white spruce, red maple, to some extent black ash) recruitment at age 30 and again, though less pronounced, after age 50. I do wonder if shrub layers prevent regeneration at all prior to age 50. It appears recruitment prior to aspen breakup (~age 50) is continuous and uninhibited. These shrubs may become a nuisance during this transition. Like advance regeneration, an intact shrub layer likely flourishes when resources increase corresponding to aspen breakup. If there is no advance regeneration to use those resources, hazel and mountain maple will no doubt capitalize. That spells trouble for subsequent regeneration.

    Thanks for the comment.
    Mike Reinikainen