By James D. Arney, PhD, Forest Biometrics Research Institute
Forest planners should be aware of a basic misunderstanding for classifying forest productivity.
Yield tables and traditional growth models have imbedded site index curves. These site curves assume a fixed trend in height growth beginning at age zero. This imbedded assumption implicitly defines the kind of silviculture which will be applied to the stand being projected. Therefore, within a given yield table or traditional growth model, there is only one yield level for each site class. This is a false assumption when viewing actual forest practices.
If any departure in silviculture is invoked, then these yield tables and traditional growth models are not representative of the stand in question.
The point being made in this discussion is that while site capacity may be constant on a given acre, the yield capacity is still highly variable due to silvicultural effects. Therefore, there is no constant value for culmination of mean annual increment (MAI) for a given acre.
For these reasons, the Forest Projection and Planning System (FPS) uses the 10-meter site index method. This method separates macro-site (2nd 10-meter log segment) from micro-site (1st 10-meter log segment). Early silvicultural treatments (brush control, planting stock, vegetation management, pest management, browse control) have a significant impact on tree vigor, growth and survival. These effects are calibrated using the FPS Cash Card methodology in young stand surveys.
10-meter site index is simply the rate of height growth in the second 32-foot height segment of a healthy tree (32-feet plus trim = 10 meters in height). See Table 1. The butt segment growth rate is more highly influenced by silviculture than macro-site capacity (soil, moisture and length of growing season). Every practicing forester has commonly observed early growth and survival differences in plantations versus natural stands. There are multiple examples in the literature for every species and region.
The magnitude of the departure of silvicultural effects within a given site index level may be observed in Figure 1. A large number of Douglas-fir trees were felled and measured across a wide range of site capacities and range of early silvicultural regimes. These regimes included natural regeneration under an overstory, natural regeneration in clearcuts and plantations with intensive vegetation control (such as progeny trials).
As may be observed in Figure 1, early height growth rates are not constant as assumed in traditional published yield tables and site index curves (King, 1966; Monserud, 1985; and Wensel etal, 1986).
The next question becomes “How influential is early height growth and survival on yield capacity within a given site class?”. The FPS growth model was used to generate 100-year yield projections for each of five traditional site classes – 50, 60, 70, 80 and 90 feet at breast height.
For each site class the FPS Cash Card was used to generate three levels of early silvicultural treatment response in growth and survival. These three levels are well within observed values.
All yield projections were initiated at 300 trees per acre in FPS merchandizing trees in 16-foot long segments to a 5-inch minimum scaling diameter and assuming Scribner log rule.
It is quite obvious that yield capacities vary with both site capacity and silvicultural regime. There is no constant culmination of yield per site class as is commonly referenced in the US Forest Service Manual. Yield capacity is highly sensitive to silvicultural regimes. Traditional yield tables and growth models with imbedded site curves (height/age models) are obsolete. They should not be used to evaluate alternative yield capacities due to either site capacity or silviculture.
Notice the range of yield by silviculture overlaps the yield differences between site classes.
Site index is constant, regardless of silviculture, for any specific location in the forest if measured using the 10-meter site classification method. No tables or equations are needed. Simply measure the number of years to grow the 10-meter segment length in the 2nd log of a healthy standing tree. Use Table 1 to identify the site index (Site = 100 / (#years in segment).
 Traditional growth models include stand models and Distant-Independent growth models (Monro, 1974). Distant-independent growth models include Prognosis (FVS), ORGANON, Cactos and Cryptos in the West.
King, James E. 1966. Site Index Curves for Douglas-fir in the Pacific Northwest. Weyerhaeuser Forestry Paper No. 8. Centralia, Washington. 49pp.
Monro, Don. 1974. Forest growth models – a prognosis. In: Growth models for tree and stand simulation. Royal College of Forestry. Department of Forest Yield Research. Research Note No. 30. pp. 7-21.
Monserud, Robert A. 1985. Height Growth and Site Index Curves for inland Douglas-fir Based on Stem Analysis data and forest habitat type. Forest Science 30(4):943-965.
Wensel, Lee C., Peter J. Daugherty, and Walter J. Meerschaert. 1986. Cactos User’s Guide: The California Conifer Timber Output Simulator. Agricultural Experiment Station. University of California. Bulletin 1920. 91 pages.