Habitable Treehouses: Building standards July-August 2000
Not as Simple as Swiss Family Robinson
I was a citizen/bystander but interested in alternative construction when Josephine County, Oregon, and Michael Garnier-owner, designer and builder of the Out 'n' About Treesort and Treehouse Institute-Engaged in nearly eight years of legal wrangling. I had spent six years on the International Conference of Building Officials' (ICBO) Research Committee (now called the Evaluation Committee) and maintained an active interest in alternative construction techniques and procedures while serving as Building Official of the City of Medford for 20 years. Upon hiring on as Building safety Director of Josephine County, I applies the newly coined department mission statement, "help people do what they want to do within the rules, codes and regulations," in order to clear up a lot of past acrimony and meld a working relationship with all of our citizens. Given the prior history of this project, with stop-work and demolition orders previously issued and avoided through legal maneuvering, I decided to see if a rational analysis and remedial corrections could be used to authorize the four existing treehouses with an extensive list of corrective measures. -David Bassett, P.E., C.B.O.
The desire to create uniformity in the design of structures is admirable. Uniformity allows the general population to have a high degree of confidence that there will be no surprises associated with the use of an apartment building, residence or commercial building. Ironically, this "sameness" also drives people to seek experiences different from the norm-in some cases, very different. A bed and breakfast facility employing treehouses provides an opportunity for those wishing to have such an experience. It also, however, creates unusual problems for the engineer, owner and local building officials. This is the story of the Out 'n' About treesort and Treehouse Institute, located in Takilma, Oregon.
Alternate materials and methods of construction are anticipated and allowed for in commonly used building codes. The use of trees as a foundation and as a system of vertical cantilevers to support structures stretches this concept to its limits. When designer and builder Michael Garnier first asked this engineer for help, he was advised that it was unlikely that his treehouses-and the trees that supported them-could be shown to meet the requirements of the State of Oregon 1990 Edition Structural Specialty Code, the operative code at that time. It would be an understatement to say that Garnier was unusually persistent, and it was eventually agreed that an attempt would be made to codify these structures.
After considerable preliminary work, a proposal was made in March 1995 to Frank Hunnicutt, the local building official at that time that outlined a program including the following steps:
- One hundred-plus percent vertical live load testing. Since Garnier had managed to find a Loophole allowing him to open for business even while being the focus of ongoing legal action, these tests were completed prior to the proposal in order to establish a basic degree of confidence.
- The buildings and all the attachments were to be analyzed in a conventional manner.
- Photographic analysis was to be conducted that would allow determination of leaf area and other parameters necessary to predict actual stresses in the trees. An exhaustive literature search was conducted to obtain applicable information. The Oregon State University Forest Engineering Department was able to provide information regarding the use of standing trees as structural members in high-line logging systems. Technical research of the load transfer and load shedding capabilities of living structural systems is still in its infancy, but some approximations were available. Specific wind drag values were adopted, and the resulting total forces and stresses were computed.
- Determining an appropriate stress level for living trees was a separate problem. Since the trees involved in this study had reached full maturity a hundred years earlier, extensive U.S. Forest Service research conducted at the turn of the last century was deemed appropriate for use. The methods and assumptions used then to reduce test values to design values remain the basis for those employed today (modified as appropriate to take into account the declining quality of second growth timber).
- On-going inspection of the trees, interconnections and structures themselves was agreed upon. Specifications for these tests were included in the plans and specifications adopted. Work on this project took place over eight years. Continuing operational history was very important to the gathering of site-specific data. Authoritative technical literature stated that living structural systems are able to equalize and limit stresses, and it was felt that such an unusual capability needed to be verified on-site. The passage of time showed that the trees in question did indeed grow additional material in the areas of highest stress.
The resulting "Peacock" treehouse is 18 feet (5486 mm) above the ground at its first story floor, covers just 99 square feet (9m ), and is used as a detached bedroom with only limited amenities. Combined vertical and horizontal stress in the main cantilevered tree limb were calculated to be 3,356 psi (23 139 kPa) in an 80 mile-per-hour (129 km/h) wind and the derived allowable stress was 2,546 psi (17 554 kPa). This produces a design stress of 2,517 psi (17 354 kPa) when the actual stress is adjusted by 133 percent, as allowed by the prevailing code. It is important to note that the structure is not occupied during windstorms.
In the meantime, Garnier had proceeded with the construction of yet another treehouse at Out 'n' About, this time supported by 13 small trees, and began constructing treehouses in other jurisdictions and even other countries. A program was undertaken to develop a special fastener that would have superior properties when used to support loads in living trees. The result is a specialized lag bolt with integral large diameter shear/compression washers and is conservatively rated at 4,00 pounds (1814 kg) of load.
Building officials faced with this or similar challenges may gain some insight form the experience of the Josephine County Building Department and others involved with the Out 'n' About treehouses. As with carnival rides and theme parks, such facilities are by their nature more dangerous than those commonly evaluated by local building officials. Cooperation among interested parties is critical to reaching a mutually acceptable conclusion. The following sequence of steps should assist in the establishment and maintenance of a cooperative environment:
- Identify common ground, prioritizing the use of standardized code-accepted design principles wherever possible.
- Identify remaining areas of dispute in a document acknowledged by all parties and establish a work plan that clearly defines the approach and assumptions to be used in subsequent analysis.
- Conduct live load tests and resolve any unsafe conditions.
- Determine if continued occupancy is acceptable with appropriate restriction
- Re-inspect critical members and connections frequently.
Technical Discussion of Treehouse Analysis
The superposition of building loads on a living structural system created a difficult analysis problem. Living structural systems attempt to achieve a state of uniform stress by growing additional solid material to support additional loads (Mattheck, Bethge). This implies that, after an appropriate length of time, the tree itself will compensate for applied external loads. It is not advisable to allow the tree limb area to be trimmed as a trade off against the projected area of the structure due to the large difference in drag characteristics. In addition, site-specific characteristics must be well understood because they play a major role in the ability of the trunk and root system to resist de-lamination and uprooting. Little definitive information was available at the time regarding the development and distribution of stresses in living trees, but a thorough search of relevant technical literature did provide some guidance. The actual value of wind loads transferred from the leaf mass to the relatively flexible tree structure is of particular importance. Speck, Spatz and Vogellehner state that 50 percent of the projected area acting as a flat plate will approximate the wind load. This may be determined by enlarging high-definition photographs to exact scale and then measuring the actual leaf area. It is certain that increasing wind velocity causes the leaves and stems to shed load once they are no longer able to resist wind loading. The projected area is ultimately reduced to that surface area described by the trunk, limb and stem outline. Thus, the wind load caused by the leaves themselves reaches a maximum value at lower wind speeds and remains limited or actually drops (leaf stem failure) with the increase of wind velocity.
The wind loads on the remaining projected area continue to climb with increased wind speed, but there is considerable non-linearity since the highly flexible portions of the tree structure are limited in their ability to transfer increasing bending moments into the more ridged portions. As the wind speed continues to increase, the trunk either delaminates or the entire root system is ripped from the ground.
Examination of nearby trees similar to those to be used in treehouse construction will reveal the capabilities of the specific species at that site to resist such failure. Technical articles state that living systems approach a safety factor of 50 percent, which may explain the robust anecdotal performance. Some safety factor is initially consumed by the addition of the treehouse loads but, as noted previously, trees will reconfigure to achieve the original safety factor after three years.
The engineering solution consisted of calculating the total applied moments at the ground plane based upon an estimate of the probable wind loads acting at the center of the wind drag areas, multiplied by the appropriate moment arms. The results of similar calculations of the wind drag (projected area method) of the structures multiplied by the appropriate distances was then added to the above moments. The stress in the tree with and without the structures was thereby derived. Seismic loads are proportional to the mass-induced inertial forces and should be minimized by utilizing lightweight frame construction. That is, wind forces govern by a factor of 2 to 1.- Charles S. Greenwood, P.E.
SummaryBased on rational analysis, the Josephine County Building Safety Department and Charles S. Greenwood, P.E., worked closely to identify and correct structural deficiencies and achieve code compliance. The final inspections addressed handrails, guardrails, stair rise and run, smoke detectors, and a myriad of specific structural considerations.
Charles S. Greenwood, P.E.,
has operated his own consulting business for the past 20 years. His practice includes stress analysis and energy system analysis and design. An extensive technical library put together by his late father, also a consulting engineer, provided pertinent historical information for the analysis and derivation of appropriate stresses in living trees. He and his wife are long term residents of the Illinois Valley in Southern Oregon and have four children. Greenwood graduated from California State University at Sacramento in 1968 and holds patents in several high-tech fields. He is also a partner in XYZ, Inc., a product development company specializing in computer modeling, analysis, graphics and tool design.
David A. Bassett, P.E., C.B.O.,
has been a building official for 25 years. He graduated from Oregon State University with a B.S. in mechanical engineering in 1968 and, after returning from a position at Edwards Air Force Base in California, completed his M.S. in 1971. Bassett served as 1985-1986 ICBO President/Chairman of the Board and is presently the Building Safety Director of Josephine County, Oregon. He and his wife, Kathleen, are looking forward to retirement and enjoying their vacation home on the coast.
U.S. Forestry Service Circular #213, Presented in Mechanical Engineers' Handbook. Lionel S. Marks, Editor in Chief, First Edition, Seventh Impression, 1916. "Unit Stresses in Structural Materials, a Symposium." Transactions of the American Society of Civil Engineers, Vol. 91, 1927. J.A. Newlin, U.S. Forestry Service Forest Products Laboratory. "Structural Timber." General Engineering Handbook. O'Rourke, Second Edition, 1940 (cites tests conducted by U.S. Forestry Service Forests Products Laboratory, Tech Bulletin 479, U.S. Department of Agriculture).
Two values for fiber stress in bending at rupture were obtained for White Oak in green condition form tests conducted over a 40-year period by the USFS Forest Products Laboratory. Their research shows comparable specific gravities (density) are directly related to comparable strength values. The lower of the two values (8,160 psi [56 261 kPa] and 8,300 psi[57 226 kPa] )multiplied by the 31.2-percent initial de-rating factor indicates a basic design bending stress of 2,546 psi (17 554 kPa). Therefore, the stress diagram (figure 1) was developed to determine those locations, if any, where the maximum allowable stress was close to actual applied stress under full load conditions. The conclusion is that the trees were well within their structural capacity, except as shown and addressed by remedial corrective measures.
"Stabilities of Plant Stems with Strengthening Elements of different Cross-Sections Against Weight and Wind Forces." Contributions to the Biomechanics of Plants. Speck, Spatz and Vogellehner, 1989.
"Tree Form, Height Growth, and Susceptibility to Wind Damage." Acer Saccharum. King, 1985.
Gale Damage to Amenity trees. Burdekin, 1978 (suggests that uprooting is more common in tall oaks and that fungus is a major factor for all).The Mechanical Design of Trees. McMahon. (discusses growth regulating hormones' impact on secondary cambium growth).
Tree Branch Angle: Maximizing Effective Leaf Area. Honda, 1978.
The Theory of Tree Bole and Branch Form. King and Loucks, 1978 (says that elastic instability is near n=1.6 for small and medium trees; also, that high winds cause flexing, which reduces total bending moment).
Effect of Mechanical Stress on Growth and Anatomical Structure of Red Pine: Torque Stress. Quirk, smith and Freese, 1975.
An Approximate Analysis of the Momentum Balance for the Airflow in a Pine Stand. Bergen, 1976.
Transfer Processes in the Plant Environment. DeVries and Afgan, 1975.
"Transport of Micronic Particles from Atomsphere to Foliar Surfaces." Belot and Gauthier. Heat and Mass Transfer in the Biosphere. DeVries and Afgan, 1975 (discusses the ability of leaves to hold microparticles in winds).
Simulation of Flow Above forest Canopies. Sadeh, 1974.
Failure of trees Induced by Delamination. Mattheck and Bethge, 1991 (load-bearing biological structures always try to grow into a state of constant mechanical stress).
Uprooting and snapping of Trees: Structural Determinants and Ecological Consequences. Putz, Coley, Lu , Montalvo and Aeillo, 1983 (says that wood flexibility and tree flexibility are not statistically related).
Root Architecture and Tree Stability. Courts, 1983.