ORGANON is an individual tree growth model developed for Southwest Oregon, Northwest Oregon, the lands of the Stand Management Cooperative, and red alder plantations in Oregon and Washington. It will project stand development for several species mixes, stand structures and management activities.
HISTORY OF THE DEVELOPMENT OF ORGANON
By
Dr. David W. Hann
January 31, 2011
The first version of ORGANON was developed in southwest Oregon as part of the Forestry Intensified Research (FIR) program. The project was ranked the highest priority amongst silviculturists in the region. The original intent was to develop a new variant of the PROGNOSIS model (as it was called back then) and, as a result, Al Stage and others from his project visited southwest Oregon to support my proposal and to explain to the silvilculturists the basics of his PROGNOSIS model. The advent of the IBM PC changed the intent. I decided that the model should be run on a PC for maximum accessibility. Therefore, I decided to design my own PC based software to run the model and the PROGNOSIS people insisted that their model could run only on a mainframe computer. This decision was re-enforced after visiting the University of California, Berkeley to see what Wensel and Biging were doing with their growth modeling work. They too were designing software to run their model on a PC. One area that I differed from them was in using the PC to estimate the parameters of the models. Wensel and Biging thought this could only be done on a mainframe, while I decided that the availability of statistical packages SPSS and BMDP for the PC indicated there were no disadvantages to estimating the parameters on the PC. I therefore modified Grosenbaugh’s REX linear regression program to run on the PC so that I could take full advantage of its powerful screening capabilities. I also took Ratkowski’s nonlinear regression software and modified it so that it could estimate parameters for data sets/number of parameters too large for SPSS and BMDP.
The decision not to develop a new variant of PROGNOSIS freed me from having to use the exact model forms used in PROGNOSIS. I therefore chose to evaluate all available model forms available at that time as I developed the equations in ORGANON. These model forms came from the PROGNOSIS, CRYPTOS, CACTOS, SPS, and STEMS (the eastern version of FVS) modeling efforts going on at that time plus any others found in the literature. As I developed the ORGANON equations, I was able to improve upon each one. Working with me on this variant were Doug Maguire (my PhD student who went on to serve as a professor at the University of Washington, the University of Maine, and is now the Giustina Professor at Oregon State University), John Scrivani (my PhD student who went on to serve as a professor at VPI and is now working for the state of Virginia), Dave Larsen (who earned a PhD at the University of Washington and is now department head at the University of Missouri, Columbia), Martin Ritchie (who worked for me after earning his MS with me, went on to get his PhD with me , and is now a project leader with the USFS Pacific Southwest Experiment Station and is the author of the CONIFERS young stand model), Chao-Huan Wang (a PhD student of mine who is now a professor in Taiwan), Abdel Azim Zumrawi (a PhD student of mine that now works for the British Columbia Ministry of Forests and has an affiliated appointment at UBC), Dave Walters (who earned a PhD at the University of Minnesota and now works for Roseburg Forest Products), and Merlise Clyde (who earned a PhD in statistics). The work involved developing equations for predicting the following: (1) total stem volume, merchantable stem volume, taper, stump diameter (all three incorporating CR), bark thickness, branch diameter up the stem (for estimating wood quality) and height growth rate for the six major conifer species in the region; (2) H/D, MCW, LCW, HCB, diameter growth rate and mortality rate equations for all 18 tree species measured on the plots; and (3) maximum size-density trajectories for stands in the region.
At the same time that I was developing the southwest Oregon version of ORGANON, I also was using data from the new College of Forestry’s inventory that I had designed to develop the northwest Oregon variant of ORGANON (it originally started as the Western Willamette Valley variant but validation work indicated its use could be expanded to Northwest Oregon). Martin Ritchie did the early basal area increment equation and the early height increment equation as part of his MS thesis. Others working on this version were Chao-Huan Wang and Abdel Azim Zumrawi.
The Stand Management Cooperative at the University of Washington chose me develop a new variant of ORGANON (SMC-ORGANON) using the permanent research plots currently available at that time in the region (southwest B.C., western Washington, and northwest Oregon) and what SMC data old enough to be useful for modeling. The research plots covered thinning and fertilization treatments so I was able to develop treatment response modifiers for thinning and fertilization. These were done in a manner that could then be applied to the SWO and NWO variants of ORGANON. Working with me on this project was Dave Marshall (he had earned a PhD in Quantitative Silviculture under Susan Stafford at OSU, went to the USFS PNW Station to replace Bob Curtis after Bob retired, and then was hired by Weyerhaeuser Company). Also working for me was Mark Hanus (who was working on his MS under me at the time developing the stand visualization program VIZ4ST; he went on to earn a PhD under me examining a new approach for characterizing the light environment using spatial information of the trees, topography of the stand, and the VIZ4ST characterization of the tree’s stem, branches, and leaves;, and he is currently working in the private sector).
At the same time that the SMC project was started, the USDI Bureau of Land Management (BLM) contracted with me to revise SWO-ORGANON so that it could be applied in older conifer stands and hardwood stands that were being used by the northern spotted owl. The original variant of SWO-ORGANON was applicable to conifer dominated stands of up to 120 years of age. This work involved sampling stands with tree ages of 350+ years of age, and stands with over 70% in hardwoods. Working with me on the model building part of the project was Mark Hanus (as he worked on his PhD). This revision required another extensive sampling effort in southwest Oregon.
After more than 10 years of additional data collection on the SMC installations, the SMC contracted with me to develop revised equations for SMC-ORGANON. Working with me was Dave Marshall when he worked at the USFS PNW Station. At this time, I also worked with Peter Gould, Dave Marshall, and Connie Harrington of the PNW Station to develop improved H/D, HCB, diameter growth rate, and mortality rate equations for Oregon white oak and to develop genetic gain multipliers for the height growth rate and diameter growth rate of Douglas-fir. Also at this time, I was involved with Doug Maguire to develop height growth rate and diameter growth rate modifiers for characterizing the impact of Swiss needle cast. All of these efforts have been incorporated into ORGANON.
The most recent ORGANON modeling work has been to develop a new variant for red alder plantations (RAP-ORGANON) for the OSU Hardwood Silviculture Cooperative (HSC) using data from research installations in western Washington and western Oregon. This work was funded in part by the USFS Pacific Northwest Station’s Agenda 2020 program and received their highest priority ranking (the 2020 program is a joint effort of the PNW Station and region industry) Because of the quality and breadth of the data, this work has allowed the development of the most comprehensive, coordinated set of equations yet, including equations for predicting: top height, H/D, MCW, LCW, crown profile, HCB, diameter increment, height increment, crown recession rate, mortality rate, size-density trajectory, and thinning multipliers for the dynamic equations. Dr. Aaron Weiskittel (Professor at the University of Maine) worked with me on the development of the top height equation that was published in 2009. Andrew Bluhm and Dr. Tzeng Lam assembled the basic data sets used in developing the model and Andrew Bluhm assisted in fitting the equations.