Keynote Comments by Jerry F. Franklin

Managing Young Stands to Meet LSR and Riparian Objectives

Workshop held at Portland, OR on August 29, 2001

The Northwest Forest Plan (NWFP) provides for silvicultural activities to accelerate development of old-growth forest attributes in young stands (<80 years old) located within Late Successional Reserves.  I have been asked to provide some perspective on the intentions of the Forest Ecosystem Management Assessment Team (FEMAT) when this provision was conceived and incorporated into Alternative 9 of the FEMAT report.  I am happy to share my recollection of the original rationale for this provision with you as well as my views on scientific considerations relevant to development and adoption of guidelines and prescriptions to implement the provision.

My comments are focused on the westside and high-elevation forests, which are periodically subject to intense stand-replacement wildfires, unless otherwise noted.  Stand management problems and appropriate silvicultural practices are much different on the low to middle elevation eastside forests, which were historically subject to frequent, light- to moderate-intensity wildfires.

Considerations in the FEMAT Process

Large Late Successional Reserves (LSRs) (or, sometimes Habitat Conservation Areas or HCAs) were incorporated into most FEMAT alternatives, including Alternative 9.  The concept of and justification for maintaining large, contiguous blocks of late-successional forest habitat first surfaced publicly in the Northern Spotted Owl conservation plan in the form of Habitat Conservation Areas (HCAs) (Thomas et al. 1990).  However, intact blocks of late-successional forest no longer existed at the desired scale so the HCAs of the owl plan and, later, the LSRs of the NWFP actually incorporated fragmented landscapes, portions of which had been logged and planted.

The FEMAT scientists were aware that LSRs included a lot of planted clearcuts occupied by the uniformly dense stands of commercial conifers created as a part of timber management programs.  Furthermore, these stands typically lacked significant legacies of large, old trees, snags, and logs from the preceding natural stand.  Consequently, the stands were structurally simplified as well as highly homogeneous.

The FEMAT scientists concluded that:

• Structural conditions differed markedly between the plantation stands and the natural young stands that developed following wildfire; consequently,
• Plantation stands were likely to develop on different and, perhaps, slower trajectories than those followed by existing late-successional forests; and
• Active management of plantation stands could help move developmental processes closer to rates and paths followed by natural stands.

And, if the active management provided some merchantible timber in the process, that would contribute to the economic objectives of the plan.

The FEMAT team debated guidelines for management of young stands in LSRs.  One issue was whether the upper age limit of stands should be 50 or 80 years and another was whether to extend management opportunities to all young stands or just plantations.  Ultimately the decision was to consider all stands less than 80 years old as candidates for management, with the proviso that the management activity must be judged as having a positive effect on recovery of late-successional conditions.

What are Late-Successional Conditions?

What were the late-successional or old-growth conditions that the FEMAT scientists wished to emulate?  Fundamentally, they were the high levels of structural diversity found in late-successional forest.  But, what does “structurally diverse” mean?

First, structurally diverse means a rich variety of individual structures.  This includes a variety of tree sizes, conditions, and species--including some large, old trees with their individualistic canopies, decadence, and large branch systems. Other important structures are large snags and large logs on the forest floor, represented in a variety of decay states, and well developed forest floors.

Second, structurally diverse means a high degree of spatial variability in structure in both the vertical and horizontal dimensions.  The vertical heterogeneity includes the presence of a canopy  (green foliage) that is essentially continuous from the ground to the tip of the tallest crown, albeit not at every point within the stand.

Horizontal heterogeneity means a high degree of spatial patterning within the stands visible as structural patches, including canopy gaps (openings) and areas with high stem densities (sometimes labeled as “anti-gaps”).

How Did These Structural Attributes Develop?

Time is certainly an important factor in the development of the structural diversity characteristic of old-growth forests but there are many specific processes at work. Franklin et al., 2001 provide a comprehensive review of structural development in natural Douglas-fir forests.  Some processes, such as the growth and development of decadence in the live trees and processes of decay in snags and down logs, are obvious—and time dependent.  Other processes are not so obvious.

Developing complex canopies (vertical diversity) involves several processes.   Most prominent is the establishment and growth of shade-tolerant tree species (e.g., western hemlock and western redcedar) into the middle and upper canopy levels.  Establishment of shade-tolerant species can be slow, actually limiting the rate at which stand development occurs, as observed in many existing mature (150 to 250-year-old) natural stands.  Also important, but not as prominent, is regeneration of canopy on the mid- and lower stems of dominant Douglas-fir in the form of epicormic branch systems, generating a deep live crown and creating massive canopy “platforms” important to wildlife and epiphytes.

These and other processes not only result in foliage continuity from ground to top-of-crown but also re-distribute it vertically—the forest canopy evolves from being strongly “top-loaded” when young to strongly “bottom-loaded” when old.  This shift in foliage distribution is something that is not well known despite its importance ecologically.

Developing complex spatial patterning or structural patches within the stand (horizontal heterogeneity) is largely the result of patchy or spatially aggregated mortality caused by diseases, insects, and wind.  Forest “gaps” are an obvious manifestation of this patterning.  In dense young stands, competitive mortality among trees (“self-thinning”) dominates.  Important features of competitive mortality (from a stand development perspective) are that it (a) results in death of smaller trees and (b) has a relatively even spatial distribution—i.e., it creates uniformity in the stand.  Mortality in older stands is dominantly the result of diseases, insects, and wind.  These agents (a) kill large trees as well as small and (b) tend to be “contagious”—i.e., they make holes in the stand!  Mature (80 to 200-year-old) stands often represent transitions between these two patterns of mortality, as pests, pathogens, and wind gradually become more important than competition.

Traditionally, forestry has recognized only competitive mortality as being a normal and acceptable part of stand development, in part because foresters have only managed young stands.  Mortality generated by pests and pathogens, which removed large (commercial) trees and produced holes in stands, has not been viewed as an acceptable and normal part of stand development.  Indeed, traditional thinning practices were specifically designed to capture much of the competitive mortality of the young stand and to forestall the aggregated mortality, such as that resulting from the natural and expected outbreaks of Douglas-fir bark beetle in mature and early old-growth stands.  Forests with such outbreaks were viewed as “unhealthy”!

In closing this section, it should be very clear that developing old-growth attributes is not just about creating stands with uniformly-spaced, large diameter trees!  Life would be much easier for all of us if it were!  Rather, it involves creating a multiplicity of individual structures—and heterogeneous spatial patterns of those structures.  Creating large trees is important but so is establishment and growth of shade tolerant species and creation of decadence in its many forms.

Did Old-Growth Stands All Follow the Same Developmental Path?

There is a tendency to assume that all of our old-growth Douglas-fir forests have followed the same developmental pathway.  Perhaps this is because several studies have shown similar wide ranges in ages of dominant Douglas-firs (Franklin and Hemstrom 1981, Tappeiner et al. 1997, Poage 2000).  Possible explanations for these wide age ranges include extended establishment periods due to limited seed supply, repeated disturbances, and competition from shrubs and herbs or, most likely, all of these.  Many of these stands are approximately 500 years old, having regenerated following a massive disturbance episode around 1500 A.D. that may have affected as much as 2/3 of the region.  These 500-year-old stands dominate our population of old-growth forests so it is not surprising that we tend to base our characterizations of old growth on this cohort of stands!

In fact, there appear to be multiple pathways for development of old-growth characteristics in Douglas-fir forests—as well as within a contiguous patch of old-growth forest.  Linda Winter (2000) has shown that even some of the 500-year-old stands exhibited rapid development of closed Douglas-fir stands, yet still developed the classic old-growth structural attributes.  An important age class of old-growth forests developed following a region-wide disturbance episode around 1700 A. D., such as in the Clackamas and Brietenbush River (OR) drainages.  Although there have been few age structure analyses in these 300-year-old stands, Douglas-fir establishment occurred within just a few decades in at least some of them (e.g., see Franklin and Hemstrom 1981).

Many mature natural Douglas-fir stands, such as some established following wildfires around 1840-50 and in 1902, appear to have fully regenerated within one or a few decades and to be on appropriate developmental trajectories.  Of course, others are not and some of those provide examples of what we probably want to avoid creating over large areas—i.e., 150-year-old stands with a low density of sound, large-diameter Douglas-firs over a dense understory of vine maple, salal, Oregon-grape, etc., and not much else in the way of structure and diversity.

My current working hypothesis is that there are many developmental pathways that ultimately produce the structural complexity of old-growth forest.  Furthermore, variability in pathways is probably the rule within stands as well as between them, a consequence of the complex burning patterns typical of large wildfires.

How Can We Help Development Along?

At this point I want to state clearly that I strongly support the view that well-designed silvicultural activities in the young stands regenerated following logging can accelerate the development of many old-growth structural attributes and is worth doing.  Stands on logged sites lack some important attributes of stands that regenerated following natural disturbances and we can help move them to more natural trajectories--even though these “deficiencies” were viewed as advantages at the time that we were creating these stands!

So, how can we help stand development processes along, recognizing that we are often trying to compress processes that take several centuries into a few decades?  How can we contribute to the creation of richness in individual structures and heterogeneity in the spatial arrangement of these structures?

Thinning young stands is one important tool—but much more complex thinning regimes than those designed for commercial timber production.  While we want to stimulate development of some large trees, we have many other goals.

Variable density thinnings are going to be the most appropriate general approach in my view.  These will incorporate “skips” (areas with no thinning) and “gaps” (areas that are heavily thinned) along with a dominant stand matrix that receives intermediate treatments.  The appropriate scale for the pattern in variable density thinning is still being discussed; Andy Carey and I feel that a variable scale of 0.25 to 1.25 acre is a useful starting point.  Some variability is important in both skips and gaps.  Many of us currently feel that essentially all gaps should retain at least some trees; they should not be viewed or described as “clearcuts”.   In addition to dominant and co-dominant trees removed in creating “gaps”, some others will probably be removed to stimulate growth of desirable understory plants, such as healthy saplings of shade-tolerant trees, such as hemlock and cedar, and to sustain hardwoods.  (Note: Do remember to consider the sun angle in marking to release understory that sun angle.)

Uniformly spaced thinnings from below (smallest trees only) will rarely be appropriate over more than a few acres.  Spatial heterogeneity is the general rule in natural stands and one of the great contrasts with our artificially regenerated stands.  Further stimulating stand homogeneity therefore seems inappropriate.

Thinning intensity is a critical issue; many of us are concerned about proposals for extensive thinnings to very low residual stand densities (30 to 50 trees per acre [tpa]).  Some of the natural mature stands that developed under relatively low stocking levels provide examples of what can happen—large, well-spaced Douglas-firs over a dense understory of aggressive, clonal shrubs and herbs and not much else.  Very heavy thinnings in young stands can generate such outcomes or (as mentioned by Jan Henderson at the conference) stimulate regeneration of a dense new cohort of trees, creating a two-storied stand—and not much else!  Neither of these outcomes contributes to our goal of accelerating development of old-growth structure.  Furthermore, retained trees will suffer mortality—sometimes high levels of mortality--so there is the potential to end up with very few large live trees after a century—and we are supposed to be creating stands that will persist for many centuries.

Hence, widespread application of high-intensity thinnings—thinnings of 30 to 50 tpa—does not seem appropriate.  Creation of patches at such retained tree levels does seem appropriate within a denser stand matrix.  Could we even call such patches the “gaps”?  There may be a consensus on this issue among silvicultural researchers; John Tappeiner specifically stated at the conference that he does not endorse general application of 50 residual tpa prescriptions.

Many other silvicultural activities, in addition to thinning, can help accelerate development of structural complexity.  1) Felling trees to create woody debris and killing trees to create snags; many of us feel that the best strategy in creating snags is to leave some live crown behind so that trees die gradually and decay more naturally.  2) Wounding trees to stimulate decay and cavity creation and to generate upper canopy complexity (multiple tops, regenerated branch systems).  3) Underplanting shade-tolerant tree species makes an important contribution to stand development in stands that currently lack a shade-tolerant component.  Perhaps the lack of shade-tolerant seed sources will not be as big a problem in our staggered-setting clearcuts as it is in some of the mature, naturally-regenerated stands where these species are absent over large areas.

Andy Carey and his “Biodiversity Pathways” model provide great insight into the many possibilities for restoring structure in managed stands.  He has advised me that the best single reference in applying this model is Carey et al., 1999.

So, silvicultural treatments can make many substantial contributions to development of old-growth forest attributes.  However, this will not be in the form of extensive “uniform thinnings from below in order to develop evenly spaced stands of large, sound Douglas-fir trees”!

Concluding Comments

I hope that foresters will approach this new challenge in the best traditions of our profession--scientifically, creatively, and humbly—and that the political institutions will support us in that process.  Surely, we have learned some lessons from the past.  Surely we do not want to lose, once again, our credibility, just as we begin to regain it by propounding professional “truths” and universal solutions.  No more mantras!

All old-growth forests did not develop from under-stocked stands.  Low stocking will not provide us with everything that we want.  It is probably not true that we must treat the plantations or most (or even many) of them will never develop old-growth forest structures.  It would take time (a long time, in many cases) but many of us believe that most of these stands would eventually get there on their own.

So, if nature will eventually do most of the job anyway—why should we proceed with young-stand treatments in LSRs?

Because, by carrying out appropriate young-stand treatments we can contribute greatly to the restoration of old-growth structure.  In my view, should do so for the good of both the forest and society.  We really do not want to wait several centuries for nature to do the job alone, assuming that she will.  Good-quality old-growth forests are in short supply in our region—we need to expand the extent of structurally-complex forests as quickly as possible to achieve our goals, including reduced risks to late-successional forest species.  We need to re-establish the integrity and capability of the LSRs as quickly as possible.

Structural restoration in young stands will require a new level of creativity and some very different perspectives on silviculture. In proceeding we need to remind ourselves that the structural development of natural Douglas-fir forests is much more complex and takes much longer than we had previously imagined.  Some of the critical processes are even events traditionally viewed as undesirable and “unhealthy”.  Heterogeneity is the rule rather than the exception—time to forget the uniformity of the industrial forest stand and landscape.  And we must remember that there are limits as to how far we can go in collapsing 400 years of stand development into 100 years!

We are going to learn a lot more about old-growth forests, stand development, and silviculture through research, monitoring and management.  I would venture that our level of ignorance about the old-growth forests and their development is greater than our level of knowledge.  But we know enough about how things were and how we have changed them to move ahead.

Literature Cited

Carey, A. B., J. Kershner, B. Biswell, and L. D. de Toledo.  1999.  Ecological scale and forest development: squirrels, dietary fungi, and vascular plants in managed and unmanaged forests.  Wildlife Monographs 142:1-71.

Franklin, J. F., and M. A. Hemstrom.  1981.  Aspects of succession in the coniferous forests of the Pacific Northwest. Pp 212-229 in D. C. West, H. H. Shugart, and D. B. Botkin (editors), “Forest succession concepts and application”.  New York: Springer-Verlag.

Franklin, J. F., T. A. Spies, R.VanPelt, A. B. Carey, D. A. Thornburgh, D. R. Berg, D. B. Lindenmayer, W. S. Keeton, D. C. Shaw, K. Bible, and J. Chen.  2001.  Disturbances and structural development of natural forest ecosystems with silvicultural implications, using Douglas-fir forests as an example.  Forest Ecology & Management (in press).

Poage, N. J.  2001.  Structure and development of old-growth Douglas-fir in central western Oregon.  Ph.D. thesis.  Corvallis, OR:Oregon State University.

Tappeiner, J. C., D. Huffman, D. Marshall, T. A. Spies, and J. D. Bailey.  1997.  Density, ages, and growth rates in old-growth and young-growth forests in coastal Oregon.  Canadian Journal Forest Research 27:638-648.

Thomas, J. W., E. D. Forsman, J. B Lint, et al.  1990.  A conservation strategy for the northern spotted owl.  427 p.  Portland, OR:USDA Forest Service

Winter, L. E. 2000.  Five centuries of structural development in an old-growth Douglas-fir stand in the Pacific Northwest: a reconstruction from tree-ring records.  Ph.D. thesis.  Seattle, WA: University of Washington.