Sustained forests; sustained profits

[ HOME | Lasting Forests Home | Table of Contents | The Finder | Glossary ]

The g-Score: Site Quality for Eastern US Tree Species

Usually, "site" is considered to be all of the interrelated soil and topographic factors affecting tree growth. That concept is pretty limited because growth is also strongly affected by growing season length, degree-days (temperature), elevation, precipitation, and the complex phenomena of evapotranspiration. Solar radiation received by the trees has been noted as a primary variable, but generally ignored because it was too difficult to obtain and assumed to be correlated with aspect.Competition between trees for light and moisture is especially keen on the better sites. Of course, tree height generally increases with age. Separating the factors that influence height-growth-to-the-maximum, then unifying them to allow an accurate, precise estimate of likely tree height at tree age 50 is one part of the role of the g-score used within Lasting Forests. This estimate seems necessary because of mast use of the tree height measure. We move away from it as rapidly as possible and concentrate on the score itself for a species. The g-score expresses our best estimate of the ability of a 10 x 10-meter tract of land (an Alpha unit) to achieve the maximum median bole weight for trees of any age over 10 years when optimally spaced.

Using modern technology and the extensive private library of Lasting Forests, a synthesis of forestry knowledge has been developed. We have combined knowledge of the key factors influencing tree growth to produce the best available estimate of site quality for prescriptive work. Site Index is a widely used index. A tree stand on a poor site may be 50 feet tall; a tree stand on a good site may be 90 feet tall. The height is the site index number. The number of logs in a tree is of primary interest to people interested in wood values (often called "stumpage"). Site index is also a measure that can be related to other factors such as the species of trees that will grow, wild animals, and understory vegetation. Site index has never been very precise. Variation has been high, but few studies have described the likely deviation or error. An site index "class", some 5 units (or feet), is usually acceptable. Beck and Trousdell (1973) reported four studies of error conducted before 1973. They thought that site index curves were biased. There has typically been an age and site bias and an assumption that there is a constant curve shape. The result is that site quality is overestimated in the younger stands and underestimated in older (greater than 50) age stands. Underestimation is widespread in available curves, apparently 10 units. Curves are species specific in shape. Rates of height growth rise rapidly on the best quality sites and then become relatively slow. On poorer sites, growth increases slowly but is maintained. There are may problems in estimating site index, including the trees selected, past site treatments, even the accuracy of aging and tree height measurement.

A measurement called FSQI has been developed with linear categories for several measures of land form that influence site quality for some tree species. We prefer to seek non-linear expressions of site quality.

Using geographic information systems, an extensive data base, the work of McCombs (1997), and the older reports of Trimble and Weitzman (1956) and Yawney and Trimble (1968) (who used curves by Schnur (1937), we created estimated site index maps for the ridge and valley province of western Virginia(Figure 1).

These maps were created for the oaks - red, scarlet, black, chestnut, and white oaks. Adjustments are being made in the equations and maps based on them as data are obtained and concepts revised. The estimates provided will equal or surpass in accuracy any now used for large area work and will embrace within one site-index class (plus or minus 10 feet) all serious estimates.

Improving rapid tree height measurements, using GPS to locate trees, and developing non-linear models will improve the site quality estimates.

The equation for the site index estimate resulting in a map of SIE 1.0 is:

SIE = [antilog [1.9702 + 0.0618A + 0.0012P - 0.0025 - 0.1509D]] + K

based on Yawney and Trimble (1968) where:

A =transformed aspect, the sine of the azimuth of slope direction from the southeast (a deviation from the conventional digression from 45°, the northeast) plus 1.0. We have concluded that the published sign of this factor (-) should have been positive as shown here. This change, however, is under investigation.

P =slope position, using the position designations developed by McCombs (1997), was translated as approximate proportion of the distance from the ridge, the slope length. The distance proportions are shown in the nearby table.

Ridge Top	0	0.25
Summit	5	0.07
Side Slope	60	0.60
Toe Slope	80	0.85
Plateau		0.70
Flood Plain	100	0.95

S =slope is expressed as a percent (e.g., where 45degrees is 100% slope)

D = the reciprocal of soil depth where... the layer used to estimate D was that for landform (McCombs 1997) and the soil depth were assigned as in the following table.

Landform	Depth in Feet
Very convex	0.5
Convex	1.5
Planar sloped surface	2.0
Planar horizontal surafce	3.0
Concave	3.5
Very concacve	4.0

Figures 2 and 3 show two areas where the site index has been calculated based on the following other adjustments.

K =moisture index (developed by Klopfer (1997)) is used to modify the results by up to 5 index points; the higher the moisture index, the greater the g-score.

C = A correction based on Beck and _______ ( ) to adjust for underestimation found in previous work and to adjust for the non-linearity in the models.

Literature Cited:

Beck, D.E. and K. B. Trousdell. 1973. Site index: accuracy of prediction. U.S.D.A. Forest Serv. Res. Paper SE-108, Southeastern Experiment Station, Ashville, NC. 7 pp.

Klopfer, S.D. 1997. Insolation, precipitation, and moisture maps for a Virginia geographic information system. M.S. Thesis, Virginia Polytechnic Institute and State University, Blacksburg, VA. 84 pp.

McCombs, J.W. II. 1997. Geographic information system topographic factors maps for wildlife management. M.S. Thesis, Virginia Polytechnic Institute and State University, Blacksburg, VA. 141 pp.

Schnur, G.L. 1937. Yield, stand, and volume tables for even-aged upland oak forests. U.S.D.A. Technical Bulletin 560, Washington, D.C. 88 pp.

Trimble, G.R., Jr. and S. Weitzman. 1956. Site index studies of upland oaks in the Northern Appalachians. Forest Science 2: 162-173.

Yawney, H.W. and G.R. Trimble, Jr. 1968. Oak soil-site relationships in the Ridge and Valley Region of West Virginia and Maryland. U.S.D.A. Forest Service, Research Paper NE-96, Northeastern Forest Experiment Station, Upper Darby, PA, 19 pp.

First draft:10/28/97

Return to the top.

This Web site is maintained by R. H. Giles, Jr.
Last revision January 17, 2000.