658 



TIMBER RESOURCES FOR AMERICA'S FtTTIJRE 



compiling the data and tabulating it. The logging 

 residue data may suffer from possible mistakes such 

 as in measuring stumps or in judging whether a cut 

 tree was in the growing stock or cull category. In 

 some instances, there is also the possibility of 

 errors in judgment and other errors in adjusting 

 timber cut data from earlier years to the 1952 

 base year. All of these possibilities were mini- 

 mized as much as possible by careful supervision 

 and checking of the work. 



Plant Residues Data From Special 

 Study 



The plant residue estimates resulted from a 

 study made especially for the Timber Resource 

 Review, the first study of its kind ever made on a 

 broad scale. The studj^ included a mail and field 

 canvass of sawmills and other plants. In many 

 regions, the canvass reached all of the larger 

 plants and only the smaller plants were sampled. 



The estimates of total plant residue volumes 

 were based, for the most part, on average ratios, 

 between residue volume and unit of product out- 

 put, applied to regional output estimates. There 

 were several recent studies which were used to 

 derive these output-residue ratios. Where no data 

 had been collected in recent years, new studies 

 were made to determine the relationships between 

 residues and output. Plant residue use, on the 

 other hand, was determined chiefly from reports 

 of firms in the various forest industries. 



Since the study was concerned chiefly with accu- 

 rate information on the proportion of plant residue 

 used, regional sampling accuracy goals were set on 

 this item. This goal was 10 percent of the propor- 

 tion of residues used, as determined for all plants 

 combined in each region ; it was met or exceeded in 

 every case. Although care was taken to avoid 

 them, nonsampling errors, such as use of improper 

 converting factors and mistakes in reporting, may 

 also have affected the plant residue data. In addi- 

 tion, an\^ errors in the timber product output data 

 were carried over into the plant residue calcula- 

 tions. 



FOREST PROTECTION 



It is recognized that present knowledge in a 

 number of fields of forest protection is inadequate. 

 Although a considerable body of data on fire losses 

 has been accumulated, there have not been enough 

 systematic surveys on losses caused by insects, 

 diseases, and animals. Furthermore, the inter- 

 relationships between fire, disease, and insects are 

 too little known. Despite these shortcomings, 

 estimates of losses to destructive agents must be 

 made in order to describe completely the timber 

 situation.^ 



' In addition to their use in calculating growth impact, 

 mortality data are also used for computing estimates of 

 net annual growth. 



The forest protection data on mortality, growth 

 loss, and growth impact represent varying degrees 

 of reliability. They are generally adequate for 

 national and regional analysis, but some of them 

 are insufficient for State appraisals. No sampling 

 errors can be calculated for these data. 



Fire Damage Data Most Adequate 



The forest protection data were assembled and 

 computed in various ways depending upon the 

 type of damage. The growth impact estimates are 

 made up of two components, mortality and growth 

 loss. Each of these components was estimated 

 separately and different kinds of procedures were 

 used in each case. Information on fire damage was 

 usually more adequate than information on other 

 types of damage. For one thing, fire damage is 

 more easily recognized; for another, standard fire 

 reports and special fire damage surveys have been 

 made for many years and in many places. At the 

 other extreme, some insect and disease damage is 

 difficult to recognize and isolate and there was rela- 

 tively little information on hand. In these cases, 

 average annual losses were calculated and used to 

 represent the growth losses resulting from 1952 

 events. 



The procedures used for estimating mortality 

 were not new; they have been used for the Forest 

 Survey and other forest inventories for many 

 years. On sample plots, all dead trees were ex- 

 amined but only those judged to have died within 

 a specified period of years were counted. The 

 ratio of dead to live volume was then determined 

 for the period and converted to an annual basis. 

 Usually, total mortalit}^ in 1952 was taken to be 

 the same as the average annual periodic mortality 

 determined in this manner. The mortality data 

 were collected along with the area, volume, and 

 other data obtained by the Forest Survey or by 

 supplementary surveys where Forest Survey data 

 were not available. 



With the total determined in this manner, 

 mortality caused by fire, and to some extent by 

 other destructive agents, was determined from 

 records and estimates relating directly to 1952 

 events. For example, the size of every fire and 

 the damage caused by it ai'e reported for all pro- 

 tected areas. Many insect and disease kills in 

 1952 were specifically known. Surveys or esti- 

 mates of these losses were usually available. 

 Mortality due to natural suppression or to causes 

 that operate over a period of years before killing 

 trees were determined from annual averages. 



Growth Loss Calculated in Several 

 Ways 



Growth loss data estimate the losses accumu- 

 lated over time as a result of the destructive events 

 of 1952. In effect they represent annual losses 



