Forest Regeneration Strategies at the Arnot

[Brett Chedzoy]

Many of you have visited Cornell's Arnot Teaching and Research Forest over the past year and noticed some big changes, namely several large regeneration harvests aimed at replacing aging stands with new ones that are as good or better as the old ones in terms of future quality and value.  For those of you who won't make it back there anytime soon but are interested in following the progress, I'll provide periodic updates here.

[Brett Chedzoy]

Some background:  Cornell's 4,200 acre Arnot Teaching and Research Forest has about 3,000 acres of managed timber.  The oldest stands on the forest originated after partial to full clearing in the late 1800's.  These stands have been managed for over 30 years through one or more "improvement thinnings".  A undesirable side-effect of these thinnings was the build-up of "interfering" vegetation in the understories, though some of this problem also arose from natural mortality due to pest outbreaks and diseases like Beech Bark Disease.  Beech brush and striped maple are the two most common species in these nuisance understories, though other native and non-native woody plants are problematic in some stands. 

Despite the growing need to replace the oldest stands while a quality seed source and silvicultural options were still available, regeneration harvests at the Arnot had been on hold since the early 2000's when multiple past attempts had consistently resulted in low-quality regeneration due mostly to the inability to control these understories and the continued presence of selective browsing by deer.

[Brett Chedzoy]

Why is the control of understory vegetation and deer impacts important in the regeneration process?  Many of you have heard this (many times) before:  Tree seedlings, even those that can tolerate more shade like Sugar Maple, need ample sunlight and uninhibited growing conditions to develop in to saplings and beyond.  Past stand disturbances - whether human or natural - usually let enough sunlight reach the ground for plants to start growing.  This is usually a mix of "good" and "bad" plants.  Chronic deer browsing of many of the "good" plants (like oak and maple seedlings) tends to repeatedly set them back, both in height and vigor.  Consequently, the plants that the deer largely ignore (examples: beech, striped maple, hop hornbeam, invasive shrubs) outcompete the other seedlings.  It's common today to see understories in maturing woodlots that are +99% "bad" plants.  

Traditional silviculture teaches that three things are necessary for successful regeneration: Increased sunlight at the ground level (disturbance); An adequate seed source (of good quality); and conditions that are conducive to seed germination and establishment of seedings (i.e. sufficient soil scarification, timely moisture and temperature, lack of pest pressure, etc.).  But the woods in which we practice silviculture today are usually full of "pre-existing" conditions that must additionally be addressed to ensure good results in the regeneration process.  Therefore, interfering understories and deer pressure must also be controlled.  This is not a unique situation to New York or even the Northeast.  The two attached publications highlight these issues for studies done in Connecticut. 

Ward%20et%20al%20%282013%29%20IPSM%20invasivedeer%20herbs.pdf

Ward%20et%20al%20%282018%29%20CJFR%20barberrydeer%20regen.pdf

[Brett Chedzoy]

What are we doing to address these issues at the Arnot?  Realizing that the challenges to regeneration weren't going to fix themselves, we decided in 2016 that we would tackle these areas with whatever was needed (and within our abilities) to implement the regeneration phase.  The first area, known as "Camp Ridge" (directly behind the lodge) was predominantly diseased Beech and decadent Aspen with a thick understory of beech brush.  Fortunately, there were still healthier species scattered throughout to provide a seed source.  Plans were made to enclose the ~ 60-acre harvest with a deer mesh fence to reduce browsing pressure long enough to grow a sapling stand.  Subsequent conversations with several foresters made us realize that building the fence would only be the beginning, and that diligent maintenance would be critical for success.  This lead us to consider other alternatives and ultimately to attempt the first deer exclosure made from logging debris and low-value trees.  The name "slash wall" stuck for these deer barriers made from pushing, piling and laying slash into solid, continuous windrows. An example can be seen below (height scale demonstrated by the Invisible Forester in the foreground)

[Brett Chedzoy]

From idea to reality  

the Camp Ridge sale was sold in the summer of 2016 to a local logging and excavation company that was accustomed to "trying new things" at the Arnot.  Approximately 25% of the stumpage value was traded in exchange for building a slash wall around the perimeter of the 60-acre regeneration harvest. Additionally, the loggers are required to cut all trees down to 5" dbh.  Most of the stems < 5" are beech, so a cut stump herbicide treatment was done pre-harvest on about half of the acreage to control the smaller beech stems. The remaining acreage will be treated post-harvest or cut and left untreated for comparison purposes.  This harvest is currently about half-completed and should be finished by fall of 2018.  All trees are being cut by hand, and the wall is being built by pushing slash with a bulldozer, dragging it with a grapple skidder, and piling where needed with an excavator. 

In March of 2017, a second large regeneration harvest was initiated on the north end of the forest by a mechanized logging crew who felt that they could use their feller-buncher (a single machine) to build slash walls.  Four months later they had completed a 74-acre block that can be seen in this video: https://youtu.be/b97tjIg9jI4  (note: the diagonal line through the middle of the harvest is a pipeline right of way)

This short video shows the feller-buncher in action:https://vod.video.cornell.edu/media/Timbco+feller-buncher+building+...  

[Brett Chedzoy]

Feller-buncher slash walls

Encouraged by the results of the completed 74-acre harvest, the same crew cut three more blocks about 10-acres in size each and enclosed them in slash walls.  The first, done in July, was a declining Red Pine stand.  The scattered Oak and Cherry trees growing in the plantation were left as a seed source, as well as some White Pine. The second harvest, done in August, is being incorporated in to the sugarbush and will provide opportunities for sugarbush silviculture research (a subject on which little research has been done to date).  For an aerial view of that harvest taken in December: https://youtu.be/qzaHqAEXcHc   The third block had a high percentage of stocking in Ash, which is expected to die in the next few years from Emerald Ash Borer.  In each of the harvests done with the feller-buncher, stems down to ~ 1" dbh were cut to "wipe the slate clean" in the understory and allow all plants to regrow on equal footing.  The hypotheses is that if this is done in addition to excluding deer during for at least five years, we can greatly reduce or eliminate the need to treat the interfering understory with herbicides.  

As a side-note of interest, these later two harvest areas contained scattered large Beech that showed no signs of Beech Bark Disease (BBD) - despite most of the Beech in these stands having died or being in poor health from the disease.  Consistent with Forest Service reports that one percent of Beech (clones) in the forest are resistant to BBD, these healthy trees were left as a seed source and for research purposes.  The diseased Beech were treated just prior to harvest with Garlon to mimimize resprouting of the BBD-susceptible clones.  The Boyce-Thompson Institute at Cornell collected tissue from these trees last summer in hopes of documenting the genetic markers for resistance and creating an inexpensive DNA test for seedlings propagated from the seed of these healthy remaining trees.

[Brett Chedzoy]

Are the slash walls working?

It will probably take at least five years before the regeneration is beyond deer browsing height (human eyeball height) and we can then evaluate the composition.  In the meantime, there are numerous monitoring activities under way such as: Game cameras; A.V.I.D. plots (Assessing Vegetation Impacts from Deer -http://aviddeer.com); and periodic reconnoiters with fresh tracking snow on the ground.  The first walk around the inside perimeter of the 74-acre harvest in mid-December revealed two sets of fresh deer tracks, which had apparently fled the scene through a small gap where the wall had been built over the top of a large Hemlock top.  The coarse branches had elevated the main trunk and other slash that was piled on top just enough to let the deer squeeze underneath.  The hole was plugged the following day and there have been no signs of deer breaching the walls since then.

[Brett Chedzoy]

The learning curve to date, and some  "Frequently Asked Questions"

I'll be adding plenty of content to this section as we continue to see what's happening and time allows to get in to the fine details.  But for now:

1. Average estimated cost of building the slash walls with the feller-buncher was $1.40 per foot.  The walls were an average of 10' high and 20' wide (the contract required a wall that was a minimum of 10' x 10' and "sufficiently dense to exclude deer" based on my subjective evaluation.  Mesh fences would cost 2 to 4 times that, plus the additional expenses of maintenance.  As Peter likes to point out: "If a tree falls on a slash wall, it just makes it better."
2. The estimated "extra" cost of cutting all stems down to ~ 1" dbh with the feller-buncher (beyond what the loggers would normally cut to harvest the merchantable trees) was about one operating hour per acre.  The contractor estimated the cost of the feller-buncher to be $200/hour.  This cost is comparable to a mist-blown herbicide application.  Unlike an herbicide treatment, the mechanical removal of the understory is less sensitive to the height and density of the vegetation, the terrain, and weather conditions.  
3. Stands with abundant poles and Hemlock seem to work best for building slash walls.  Heavily-thinned stands lacking pole-sized trees require importing material from a further distance in order to build an effective slash wall.
4. Factors like terrain, timber and ground conditions will influence the cost of slash wall construction.  The least-expensive slash wall cost (<$1/ft) was on gentle terrain with abundant "junk" trees.  The most expensive section ($3/ft) was built on steep side slope.
5. A successful slash wall project starts with good planning and a capable and willing contractor.  There's no substitute for the "capable and willing" part. 

[Brett Chedzoy]

Special contractual considerations for regeneration harvests with slash walls

One of the five regeneration harvests started in 2017 at the Arnot, the one that won't be completed until later this year in 2018 is using hand felling (cutting the trees with chainsaws vs. machines) and building the slash wall through a combination of pushing, piling and stacking with dozers, grapple skidders and an excavator.  This harvest is about half completed to date, and the quality of the work is good so far.  What I think we'll learn from this harvest, however, is that it's easier to build slash walls and eliminate the interfering understory vegetation with a feller-buncher.  

In hindsight, the biggest advantage of the feller-buncher is the ability to zip off (as one visiting forester put it: "weed whack") all stems down to < 1" in diameter.  By doing so, and keeping the deer pressure off the resulting regeneration for at least five years, we believe that the desirable seedlings can compete with everything else (beech suckers, brambles, striped maple, etc.).  However, it'll take at least a few years to get a good assessment of this hypothesis.  

Due to the limited availability of crews with feller-bunchers, I'd be reluctant to write that requirement in to the next contract until we have more experience to see what other equipment combinations will work and the comparative economics of each.  What I would incorporate in to the next contract based on our initial experiences is the following:

1. All stems down to 1" dbh must be cut or crushed to ground level.
2. Slash walls should measure at least 10 feet high to branches at least 2" in diameter.
3. The bottom five feet of the walls must be sufficiently thick and dense to prevent deer from crawling through.  This means that just pushing or laying course tops in to a windrow would not be acceptable, unless they were sufficiently interlaced as to leave no holes through which a deer could pass.
4. Wall must be fully completed within one year of harvesting the first tree.  The contract may be longer, but the point here is that one the job is started, they should keep working on it as weather permits until finished.  

[Brett Chedzoy]

Situations where simply cutting the interfering understory vegetation may not work

As explained above, the option for cutting & crushing woody stems down to < 1" dbh looks promising when compared to alternatives like a pre-harvest herbicide treatment.  However, interfering understory plants do not always come in the shape of small-diameter woody stems.  Herbaceous plants like ferns and garlic mustard, as well as some shorter growth habit invasive shrubs (multiflora rose, barberry, and even honeysuckle) would be difficult to cut and crush sufficiently to prevent them from severely competing with desirable seedlings.  Therefore, chemical or other mechanical controls may be needed for those situations.  

The picture below (compliments of Dave Jackson, Extension Forester with Penn State University) illustrates the "before and after" of fern control using herbicides.  The seedlings in the treated foreground are Black Cherry.  Once the extremely dense low shade of the fern was removed, the cherry seed was able to germinate and become established as seedlings.  

 

[Brett Chedzoy]

Gates and Monitoring

An unanticipated challenge was getting the deer out of the harvest areas, then keeping them out with secure gates at the access points.  I've asked colleague Mike Ashdown who has undertaken these tasks to share some of the learning curve on building effective gates and the monitoring technology he's been using at these checkpoints.  Lots of great pictures so far not only of curious deer (on the outside of the gates!), but every other critter known to roam the Arnot like the large blond coyote below.