| Authors: |
David Gartner, Gregory Reams |
| Year: |
2002 |
| Type: |
Scientific Journal |
| Station: |
Southern Research Station |
| Source: |
McRoberts, R.E.; Reams, G.A.; van Deusen, P.C.; [and others] NC-252. St. Paul, MN: U.S. Department of Agriculture, Forest Service, North Central Research Station: 149-153. |
Abstract
When Forest Inverltory and Analysis (FIA) changed to the national standards for the inventory system, plots with lnultiplc condition codes were introduced to the Southern Station's FIA unit. FIA maps up to five different conditions on completely or partially forested 1124-acre subplots. This change has madc producing inventory estimates more complex because the data are analyzed by condition classes (partial plot) rather than on a whole plot. Methods for analyzing by condition classes are less intuitively obvious than methods based on a single condition for an entire plot. We compared the current standard of fully rnapped plots to two methods that reduce the number of mapped conditions per plot using the following sets of rules. Rule 1 assigns the predoniinant condition to the entire plot, including plots that are partially forested. Rule 2 maps a single nonforest condition and a single forest condition, with the single forest condition predominant. The effects of these changes were shown by calculating forest area by forest type group and ownership and volume by species and diameter class. The effect of using just one condition per plot (rule 1) increased the estimated total forested area by 0.4 percent. Using rule 1 decreased the calculated total volume for the state by 3.9 percent. Using only one forested condition per plot (rule 2) did not change the total area estimate or the voluine estimates. Both methods decreased the estimated variance for the total volumc by 12 to 16 percent. The percent changes in the estimated values were greatest in the least occurring table entries, e.g., the rarest colnbinations of forest type and ownership group or colnbinations of species and diameter class.
Citation
Gartner, David; Reams, Gregory. 2002. The effects of removing condition boundaries on FIA estimates. McRoberts, R.E.; Reams, G.A.; van Deusen, P.C.; [and others] NC-252. St. Paul, MN: U.S. Department of Agriculture, Forest Service, North Central Research Station: 149-153.
