 |
 |
 |
 |
|
|
 |
 |
|
 |
|
|
 |
|
|
 |
|
|
 |
|
|
 |
|
|
 |
|
|
 |
|
|
 |
|
Study Area
The 21 counties chosen for this study were located in eastern Ohio and distributed across the Northeastern, East-Central, and Southeastern Units as defined by the U.S. Forest Service (Figure 1). Counties included were Ashland, Athens, Belmont, Carroll, Columbiana, Coshocton, Guernsey, Harrison, Hocking, Holmes, Jefferson, Meigs, Monroe, Morgan, Muskingum, Noble, Perry, Tuscarawas, Vinton, Washington, and Wayne.
These counties were selected because the 1991 FIA concluded that this area contained more than 70 percent of the state's white pine volume (Griffith et al., 1993). Further, this region is geographically well situated to supply white pine to Ohio's forest industries.
Land Classification
White pine stands were identified using four Landsat 5 Thematic Mapper images. The images utilized were as follows:
All images were georeferenced to Universal Transverse Mercator Zone 17 coordinates with the North American Datum of 1927 as the reference. The spatial or ground resolution of the Landsat Thematic (TM) data was 30 meters by 30 meters and was not resampled.
The areas classified as "wooded" by the 1994 Ohio Land Cover Inventory (DREALM-ODNR, 1994) were used to extract the multispectral data for woodland from the TM image data. This multispectral data was computer processed to develop spectral signatures for deciduous and coniferous species. A clustering program was used to group the data into unique spectral signatures that typically define land covers. The resulting signature set was reviewed, and signatures with high variability, which corresponded to multiple land covers, were removed.
A spectral distance or separability measure was computed on remaining signature pairs. If the spectral distance between two signatures was not significant, one of the signatures was not distinct enough to aid in a successful classification and the signature was deleted. The resulting signature set was used in a maximum likelihood classification program.
Each 30 meter by 30 meter pixel in the TM data was assigned into the spectral cluster it most closely fit statistically. The resulting classification was compared against ancillary data to assign each spectral cluster into an informational land-cover category, deciduous wooded or coniferous wooded. The data used to do the assignments included Landsat TM data from the spring of 1986, spring of 1987, and early summer 1988; aerial photo interpreted land-use/land-cover data from the Ohio Capabilities Analysis Program (OCAP); and National Aerial Photography Program (NAPP) 1994 photographs.
Clusters that were classified as mixed deciduous and coniferous were extracted again from the TM data, and new spectral signatures were produced and reviewed with a new classification produced for these mixed areas. All classified data were combined to produce a final "conifer" classification. A contiguity analysis of this final classification was performed to identify conifer stands three acres or larger for the 21-county study area.
Stand Data Collection and Analysis
Two stratified random samples of identified conifer stands in the study area were selected. The first was used to determine what proportion of the stands identified as conifers were white pine; the second was used to estimate white pine volume per acre and site quality.
To ensure that stands sampled would best reflect the entire geographic region, the 21-county study area was stratified into seven three-county regions (Figure 1). The proportion of stratified random samples drawn from each geographic region was determined based on the proportion of the total conifer acreage in the study area contained in that region as indicated by the spectral classification. Seventy-two conifer stands were selected to determine what percentage of the conifer stands identified by the classification scheme were white pine. Species composition of the selected stands was determined by site visit.
| ||||||||||||||||||||||||
Figure 1. The 21-county study area in eastern Ohio. These counties were selected because FIA information concluded that this area contained more than 70 percent of the state's white pine volume. The number of stands sampled for individual stand characteristics in each county is indicated in parentheses. |
Forty-one white-pine stands were selected for the purpose of estimating volume per acre and site quality. To minimize any potential effect of stand size, the distribution of stands sampled within each region was also stratified based on stand size, with the proportion of stands sampled in each size class reflecting the frequency of that size class in the region. Stand size classes used in the stratification were 04 acres, 59 acres, 1019 acres, 2039 acres, 4079 acres, and 80+ acres. The geographic distribution of the stands selected within the study area is shown in Figure 1.
Stand volume and site quality were estimated on a 0.2-acre plot randomly located within each of the 41 stands. On each plot, the diameter at breast height (dbh) of each tree, the total height of the first tree measured in each two-inch diameter class beginning at eight inches dbh, the total age of the stand, and the five-year height intercept beginning three years above dbh for three dominant or codominant trees were recorded. White pine volume per acre was estimated using an equation developed by Dale et al. (1989) for white pine in southern Ohio. For each plot, height and average dbh were used to assign board-foot volumes (International 1/4-Inch volume to a 6.0-inch diameter top inside bark) to trees in each diameter class. Site quality was estimated by calculating the 35-year site index based on height intercept equations developed in Ohio by Brown and Stires (1981).
Back | Forward | Table of Contents
All educational programs conducted by Ohio State University Extension are available to clientele on a nondiscriminatory basis without regard to race, color, creed, religion, sexual orientation, national origin, gender, age, disability or Vietnam-era veteran status.
Keith L. Smith, Associate Vice President for Ag. Adm. and Director, OSU Extension.
TDD No. 800-589-8292 (Ohio only) or 614-292-1868