5. Because of a lack of height-growth data on the uneven-aged plots, development of a 

 height-growth model was not possible. Instead, additional data needed for the development of 

 height -diameter equations were obtained from the School of Forestry at Northern Arizona Uni- 

 versity and from the Rocky Mountain Forest and Range Experiment Station at Flagstaff. 



Plot and Subplot Allocation 



With the elimination of the small subplots and those seriously affected by the highway, 

 110 uneven-aged subplots remained. Of these, 24 were reserved for validation (objective 

 number 3). To pick these, each of the four uneven-aged plots had their subplots divided into 

 low, medium, and high basal area stocking classes (one-third of the subplots in each), and two 

 subplots were randomly selected from each class. This helps to insure that the validation data 

 represent the full range of stocking conditions. 



Also reserved for validation were six even-aged plots. These were also randomly chosen by 

 selecting one plot from each of the six basal area stocking classes. The result was 86 uneven- 

 aged and 12 even-aged plots for use in developing the simulator. 



SIMULATOR DEVELOPMENT 



The development of a whole-stand simulator that can express number of trees by diameter 

 class is the main objective of this study. A 1-inch diameter class size was chosen because it 

 offered a good compromise between model complexity on one hand and sensitivity to stand struc- 

 ture and to managerial and silvicultural information needs on the other. The decision to 

 express the stand as number of trees instead of basal area in each diameter class was made 

 because many of the simulator's stand dynamic components are best expressed in terms of 

 number of trees. 



The dynamics of stand development is composed of three basic elements: growth, mortality, 

 and recruitment . 



Growth 



The process of growth takes place on all portions of the tree: roots, crown, and stem. 

 The growth of most interest to the manager, however, is stem or bole growth, which is the 

 height and diameter growth. With the lack of height-growth information in the data set, a 

 height-growth model could not be developed. Instead, height equations were developed as a 

 substitute to aid in the prediction of product potential. Lack of a height-growth model did 

 not affect the processes used to predict change in number of trees by diameter class. Such a 

 change is the difference between the number of trees growing into the diameter class and the 

 number growing out of it. The number of trees growing from one diameter class to another, 

 sometimes called upgrowth, is a function of three components: diameter growth, the distribu- 

 tion of number of trees within the diameter class, and diameter class size (Wahlenberg 1941). 



At least two approaches exist to modeling this process. In one approach, Ek (1974) chose 

 not to treat these components separately, but rather, he predicted the number of trees growing 

 into the next larger diameter class directly. Another approach is to consider each component 

 of the upgrowth model individually. This approach was used in the study because it allowed a 

 closer examination of the validity of the individual component assumptions incorporated into 

 the upgrowth model . 



4 



