APPENDIX III 



rocedures 



Commercial Forest Area 



Estimates of forest land and commercial forest land areas 

 obtained in the most recent forest survey were adopted as 

 the best available estimates as of January 1, 1963. In 

 practically all States figures were based on classification of 

 a large number of points on aerial photographs, followed 

 by field verification of a sample of the photo points. 



Timber Volume 



Estimates of timber volumes obtained in surveys com- 

 pleted within the three years prior to 1963 were adopted 

 without change. In other cases data for the most recent 

 forest survey were updated to January 1, 1963, in most 

 instances by using the growth-projection procedure de- 

 scribed below. Volume estimates for part of California 

 were based upon remeasurements of a sample of locations 

 established in the initial forest survey. 



Where necessary to insure comparability between esti- 

 mates for 1953 and 1963, information from surveys com- 

 pleted subsequent to 1953 were "backdated" to January 1, 

 1953 to_ obtain revised estimates for 1953, using the growth- 

 projection procedure described below. In the East in- 

 ventories were updated an average of 5 years; in the West 

 the average updating period was somewhat longer. 



Net Annual Growth and Mortality 



Estimates of annual growth in 1962 were based mainly 

 upon remeasurements of radial growth as indicated by 

 increment cores, using a 5- or 10-year period preceding the 

 field survey, applied to the updated 1963 inventory. 

 Estimates of mortality were based largely upon measure- 

 ments of trees determined to have died on inventory plots 

 during the 3- or 5-year period just prior to field surveys. 

 Additional allowances were made for sporadic or cata- 

 strophic mortality, where appropriate, when no evidence 

 of such mortality was found on survey plots, based on fire 

 records or other local evidence. 



Consistency Checks 



Upon completion of the updating and backdating of in- 

 ventory and growth estimates, a check was made for each 

 State to make sure that changes in inventory volumes 

 between 1952 and 1962 were consistent with differences 

 between net growth and timber cut in 1952 and 1962. 

 Inconsistencies may arise for one or more of the following 

 reasons: 



1. Sampling errors in estimates of inventory, growth, 

 cut and mortality. 



2. Estimating and reporting errors, particularly in 

 inortality estimates and in timber cut figures de- 

 rived from industrial surveys. 



3. Differences between the annual cut in 1952 and 

 1962 and the average of the 10-year period. 



In most instances inconsistencies could be attributed to 



sampling errors and annual fluctuations in timber cut. 



In some cases inconsistencies could not be explained or 



eliminated since estimates for inventories, growth and cut 



226 



in the past have necessarily been derived more or less 

 independently. This problem should be greatly reduced 

 in the future for surveys are now based on remeasurements 

 of permanent plots. 



Since estimates of inventory volumes generally are most 

 reliable, most of the inconsistencies are attributed to dis- 

 crepancies in estimates of timber cut and net growth, 

 notably the mortality component of net growth. In- 

 consistencies thus point to an underestimation of timber 

 cut and mortality, and an over-optimistic growth/cut re- 

 lationship for 1962. This did not appreciably influence the 

 long-term timber supply projections, however, for by the 

 mid-seventies, the constraining influence of rapidly in- 

 creasing stand densities became much more of a controlling 

 factor than discrepancies between growth and cut in 1962. 



A consistency check for the entire United States showed 

 a discrepancy of 17.9 billion cubic feet of growing stock, 

 or 2.8 percent of the 1963 inventory. The discrepancy 

 for sawtimber was about 54 billion board feet or 2.1 per- 

 cent of the 1963 inventory. It is estimated that not more 

 than 0.8 percent of the 2.8 percent inconsistency for grow- 

 ing stock could reasonably be attributed to sampling 

 errors in the inventory. 



Growth Projections 



Estimates of inventory volumes, net growth, and mor- 

 tality of growing stock for the period 1963-2000 were 

 based upon a stand-projection method programmed for 

 use on the Honeywell 800 and IBM 709 computers. This 

 procedure started with the updated 1963 inventory of 

 number of trees by 2-inch d.b.h. classes, derived by using 

 radial growth rates, mortality rates, cutting rates, in- 

 growth rates, and volumes per tree based on the most 

 recently completed forest surveys. 



The "potential" increase in number of trees in each 2- 

 inch diameter class in the absence of cutting and mortality 

 was then calculated using those same growth factors, as 

 described in more detail below. From these estimates, 

 deductions were made by diameter class for number of 

 trees cut, number of trees lost by mortality, and the 

 growth on trees cut and lost to mortality. These com- 

 putations in terms of total number of trees or trees per 

 acre, were made annually for the specified period. Num- 

 ber of trees were then converted to volume and basal area 

 in specified years. 



All projections were made for the area of commercial 

 forest land as estimated for January 1, 1963. Timber cut 

 in each section and region was based upon an allocation 

 of the total timber cut developed in the section on The 

 Outlook for Timber Demands. Projections were made 

 assuming no increase in the level of forest management 

 above that prevailing in 1962. 



Input factors for the computer program were developed 

 as follows: 



Number of trees per acre. — Total number of trees in the 

 inventory as of January 1, 1963, by 2-inch diameter 

 classes, and by softwoods and hardwoods, were pooled for 

 the region for which the growth projection was made, and 



