DETERMEI‘GSNG 'fHE SKY WEIGHT AND DRY VQLQME 0F GREEN ASPEN BIOL'E’S 33‘ USE C“? THE WAFER QSSPLACEMEE‘QT TECHNEQUE 7.525;: for H12 Degree of M. S. RfiCHISéE‘é' STATE UNE‘VERSETY John Grant Haygreen 1958 1mm: LIBRARY Michigan State University DETERMINING THE DRY WEIGHT AND DRY VOLUME OF GREEN ASPEN BOLTS BY USE OF THE WATER DISPLACEMENT TECHNIQUE by JOHN GRANT HAYGREEN AN ABSTRACT Submitted to the College of Agriculture Michigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Forest Products 1958 Approved 1:2&M£”M<$~ 2?; [Qoé&%ggk / / Abstract Most purchases of boltwood today are based upon either cord measure or green weight. Neither of these methods, as commonly practiced, is reliable for consistently providing information as to the actual weight of dry wood substance present in boltwood. The water displacement technique can be used to provide this information. The actual volume of boltwood can be determined by diSplacement, and this green volume can then be converted to the corre- sponding dry weight if the relation between these two values is known. In this study the ratio of green volume of boltwood to dry weight of wood substance was determined for both peeled and unpeeled bigtooth aspen (POpulus grandidentata Michx.). The variation of these ratios, 1.6., conversion factors, with the diameter of the bolts and the height in the tree was also investigated. It was found that the weight of dry wood in a unit volume of green wood was greater in bolts of larger diameter. The dry weight per unit volume remained constant, however, at different heights in the tree. Conversion factors were determined for bolt diameter classes from three inches to nine inches, and thus a factor which corresponds to the average bolt diameter of a shipment could be used to estimate the dry weight of wood in the shipment. Also determined was a factor which can be used to estimate the dry weight of wood from the green weight. Weight and volume losses which occur in aSpen boltwood under good air-drying conditions were investigated; and these were used to compare the accuracy of the two types of factors mentioned as affected by changes in moisture con- tent of boltwood after cutting. In unpeeled aspen boltwood, the green volume decreased 1.2 percent after six weeks of drying while the green weight dr0pped 10.3 percent. Estimating the dry wood substance in boltwood from the green weight appears impractical when an accurate estimate is desired. This is due to the rapid.change in weight, and thus in the true conversion factor, when the bolts are subjected to drying. Also of importance in this respect is the fact that wide variation may exist in the moisture content of freshly cut bolts at different seasons of the year. The disadvantages regarding accuracy of the green weight method of estimating could be overcome by using the water diSplacement method and the necessary conversion factor. In this study conversion factors are presented for bigtooth aSpen grown on one site in central Michigan. Two methods of obtaining conversion factors are outlined in this study, and these methods could be used to obtain factors for any Species of boltwood. DETEE'{I‘-,;1{\III=IG TTTFI DRY T lCrFJT .7de DRY VC’LULIE OF GREEK ASIER BOLTS BY ESE 0? 'IE MATTE D131 LACE—ITFIIT TF‘C’TITIQL’E by JOHN RANT HAYGREE: A THESIS Submitted to the College of agriculture Xichigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of Department of Forest lroducts 1958 ACKNOWLEDGMENTS The writer extends his sincere appreciation to Dr. Peter Koch and Mr. N. C. Higgins for their assistance and guidance during the initial planning and throughout this investigation. Grateful acknowledgment is also extended to Dr. A. E. Wylie for his help during the final stages of the study. Finally, the writer is indebted to his wife who acted as typist during the preparation of the thesis. ll ACE"::C'-IIEDG1 1:: ‘T5 0 e e e e e e e e o o e e o e e o e 0 II. III. IV. v. AIIENDIX ATTENDIK B. BIBLIOGRAIHY . . . . . . . . . . . . . . . . . . . . TABLE OF CONTELTS A 7“ ‘ N TILL‘LTIO C O O O O 0 O O O O O O 0 O O O O O O I? 1 .L IC‘UEXFS e e e e e e e o e e o e e e e e e e e V IFTTRODUCITIOI: O O O O - O O O O O O O O O O O O O l Hethods of Scaling Boltwood lurposes of the Jtudy Descrigtion of the Samrle Determination of height and Volume Losses Due to Drying Determinations Made in the Laboratory Computation of Conversion Factors RESULTS AHD AKALYSIS . . . . . . . . . . . . 20 Conversion Factors Based Upon actual Total heights and Volumes lhysicnl lroperties of Birtooth Aspen Findings Related to Conversion Factors Drying Rates of Aspen Boltwood DISCUSSIOI; O O O O O O O O O O O O O O O O O 4” C()I:CLUJIOI: e e . e e e o o e e e e e e e e e- 47 5Tlirq13T ICJIL TI‘XEIJFS e o o e 0 o- . o o o 0 A. 49 DETERMINATION O” CONVERSION FaCTORS AT ANY IICISTURE CCTITEL ‘ . . . . . . . . (D U3 ’31 (‘3') Table II. III. IV. VI. VII. VIII. IX. LIST OF TnPLFS Inysical irogerties of Bi{tooth aszen in 3amyle A, Averaged by Diameter Classes . . . . . . . lhysical Troyerties of Biptootb As:en in Sample A, Averared by Heirht Classes . . . . . . . . Definition of Conversion J-actors and the Formulas Used to Comrute Conversion Factors . Conversion Factors for Digtooth Aspen Averaged by Diameter Classes . . . . . . . .-. . . . . Conversion Factors for Bigtooth Asren Averaged by IEEig‘ht 0161 8568 O O O .1 0 O O 0 O O O O O 0 Summary of the Results of Analysis of Variance for lhysical Erogerties of Bigtooth As;en . . Results of the Kultiple Regression, Sgecific Gravity of Hood the Degendent Variable . . . . Summary of the Results of Analrsis of Variance “or Conversion Factors . . . . . . ... . . . . Weight and Volume Losses of Four-Foot Unyeeled Aspen Bolts, Samtle A, Observed Under Good Air-Drying Conditions . . . . . . . . . . . . Weight and Volume Losses of Four~Foot Peeled and Unpeeled Aspen Bolts, Sam;le B, Observed Under Good Air-Drying Conditions . . . . . . . iv 16 17 55 Fisure l. 19 LIST OF FIGLRFS Iictorial Representation of dangle A . . . . . . Gragh of the Relation of the Specific Gravity of Hood to Bolt Diameter . . . . . . . . . . . Graph of the Variation in'the Volume of Green Bark with Height in the Tree . . . . . . . . . Grafh of the Variation of the moisture Content of Bark with Peisht in the Tree . . . . . . . Grain of the Variation of Conversion Factor Number One with Polt Diameter . . . . . . . . Graph Showin? the Variation of Convers'on Factor Number Two with Bolt Diameter . . . . . Grain Shotin; Volume and Jeifht Losses in bnpeeled Asgen Boltwood, Sangle A . . . . . . Graph of the Loisture Content of leeled and Ungeeled Aslen Bolts During Air-Drying in {in ::*?OS€d LOCQtID'Q o o o o o o o o o o o O o ,n Showing weight and Volume Losses Cbserved in Both leeled and Unpeeled Bolts, Samrle B . d. I. INTRODUCTION Methods of Scaling Boltwood Most purchases of boltwood today are based upon either cord measure or green weight. Neither of these meth— ods, as commonly practiced, is reliable for consistently providing information as to the actual weight or volume of dry wood substance present. A type of measurement which consistently provides more accurate information would not only make it possible to set purchase prices more realisti- cally, but would be helpful to industry and researchers in the determination of product yield standards. The tech— nique of obtaining the volume of bolts by immersion in water, 1.6., by water displacement, has considerable prom— ise of being such a method of measurement. The possible wide variation in the amount of wood substance contained in a unit cord of 128 cubic feet is well known. Factors such as diameter of bolts, length of bolts, care taken in stacking, amount of bark, moisture content, and neatness of trimming, affect the amount of wood substance in a unit cord. Cord measurement gives only approximate information about the volume of green wood in the unit, and even less information about the volume or weight of dry wood which is present. ' 1 The measurement of boltwood by weight also poses several problems. The principal problem lies in the fact that the bolts may lose moisture, and thus weight, quite rapidly after being cut. Therefore, a conversion factor based upon a relationship between freshly-cut wood and its equivalent dry weight becomes pregressively greater in error as it is applied to boltwood in progressive stages of drying. Since the relationship between the actual weight and the dry weight of boltwood is subject to continual change after cutting, any estimate of the dry weight made on the basis of green weight is subject to considerable error unless moisture tests are made. Tests to determine the average moisture content would be necessary even if boltwood is weighed immediately after cutting since within a tree species such as aspen the moisture content may vary during the year. Jensen and Davis (7) found that the aver- age moisture content of the quaking aspen in their samples varied from 80 percent in the summer to 113 percent in the winter. sThis could mean a difference during the year of approximately 15 percent in the green weight of a unit of dry wood. Despite its limitations, the selling of boltwood on a weight basis has a number of advantages over selling on a cord basis. Three (12) lists a number of these: (1) it encourages prompt delivery which is desirable for pulping, (2) it requires no Special handling and thus saves time for both buyer and seller, (3) it provides an incentive for better piling of wood on trucks and thus increases the volume handled by the supplier. Olson (9) states that since his company began purchasing by weight in 1947 they have received no complaints on scaling. Prior to this cord scaling was used and complaints were commonplace. As stated previously, the limitations of weight and of cord measure could quite possibly be overcome by using the water displacement technique. This method would in— volve submerging loads of boltwood in a water tank and obtaining the volume of wood, or of wood plus bark for un- peeled bolts, by the change in the water level. If the relationships, i.e., conversion factors, between the volume of green wood, or of green wood plus bark, and the amount of dry1 wood are established, it would be possible to convert the volume obtained from displacement into the corresponding dry weight. The average volumetric shrinkage of bigtooth aSpen in going from a green to a dry condition is listed in avail- able tables as 11.8 percent. Thus, if a given amount of green sapen boltwood is air-dried to 20 percent moisture content the volume would deerease by only about 4 percent, lThroughout this paper the term “dry" is used to indicate wood or bark in an oven-dried condition unless another moisture content is Specified. whereas the weight could decrease as much as 40 percent. It is thus evident that the amount of dry wood present in a given amount of green wood could be much more accurately estimated throughout progressive stages of air—drying by use of a conversion factor based upon green volume than by use of a factor based upon green weight. Purposes g: the Study In order to evaluate water diSplacement as a tech— nique for determining the amount of dry mood in green bolts, information was needed pertaining to the limitations of such a system. Further information in two important areas was deemed necessary in order to evaluate this system. The primary purpose of this study was to compute factors for converting the green volume of peeled and of Unpeeled aSpen bolts into estimates of the dry weight of wood substance present. It was also possible to derive conversion factors for converting the green volume of peeled and of unpeeled bolts into estimates of the dry volume of wood present. The second purpose of the study was to compute a factor for converting green weight of aSpen into dry weight; and to compare the accuracy of this factor with the one for converting from green volume to dry weight, as affected by changes in moisture content. In order to make such a comparison the extent and rate of weight and volume losses which occur when aSpen boltwood is stacked in an exposed location were determined. The most desirable type of conversion factor would be one which gives an accurate estimate of the dry weight of wood present regardless of the moisture content of the bolts at the time of measure- ment. Another purpose of this study was to determine the influence that the diameter of bolt and the location of the bolt in the tree might have upon a single conversion factor applied to all bolts. To determine this, data were obtained on such variables as moisture content, specific gravity of wood and of bark, shrinkage of wood, and the amount of bark. All conversion factors in this study were computed from data obtained in a sample of bigtooth aSpen (Pogulus ggandidentata Michx.) which was taken in central Michigan. II. PROCEDURE Description 9; the Sample All test sections used in the study of physical prOperties were out on the same site in the Manistee Nation- al Forest in central Michigan. Bolts for the study of losses in moisture content, weight, and volume during air—drying were obtained from two areas. The first sample, which will be referred to as sample A, was cut at the same location and at the same time as the test sections mentioned in the preceding paragraph. The bolts used in the study of drying losses were cut fifty- two inches long. The ends were coated with a wax emulsion to retard moisture loss. When the bolts were delivered to the drying area a two-inch section was cut from each end. In sample A, a ten-inch section was out directly below the bottom out made to remove a fifty-two inch bolt from the tree stem. There was, therefore, a ten-inch section obtained adjacent to each of the sixty-three test bolts in sample A. These sections were those used to obtain data on the physical preperties of bigtooth aspen. The second sample of bolts used in the drying study was cut on the Allegan State Forest near Allegan, Michigan. . .oropo c3335 sun» unpoouw can use moon» no .3958 SE. 63.0325 3 canoes anon noun no "Emma: and woman Rousseau can. .4 nausea yo sofipupsomouaa :30on ..n .wrn mmmm... m M o m a . o m a. ... m. ..W n o a lei e H a I. a m B . 1.. II. 9 In; N. 1m: 0 a 1m. Ion . m H Y ... m M M lol u o m ..m_ ..N.. ,m H n m s o m at M a 3 l... 11 .....— M [en [mi m k... m m iN _ f I. .1 I. .3 In. .N In. a w o m . 0N3 no I 3.123 .3 .02 U J ......“ 42.3 ax... 2233 am rm. m Io: s m m . mm u: no ”230 55.33 I l 2: 3:33.: 52.52 3:. ..L J 1 ill .zoCoum 5.2...2: « 353.32. m e m m h m m :mm 3023.3: 3502:: :2; U nu an t (L i3. This set of sixty bolts will be referred to as sample B. No ten-inch test sections were cut from this area. These bolts were cut in the woods into eight-foot lengths, and when delivered to the drying area were cut into the desired four-foot lengths. One member of each such pair was peeled and the other was left unpeeled in order to obtain matched specimens for a direct comparison between peeled and un- peeled bolts during air-drying. Information pertaining to the sites and stands from which the samples were obtained is important as these factors define the pOpulations from which the samples were drawn. Since sample A was used as‘the basis for data on physical preperties, it is the more important sample and will be described in somewhat more detail. A total of 29 trees were cut in order to obtain the 6} bolts referred to as sample A. The average height of‘ these trees was 54 feet, and the average diameter inside bark at one foot above the ground was 7.3 inches. The aver- lage height of what was considered merchantable stem, that is the height to a 3 inch diameter, was 37 feet. These figures do not represent average figures for the stand since the trees were not selected at random, as will be explained later. These averages indicate only the size of trees from which the sample was obtained. The site index for the stand of sample A was esti- mated at 68 using the site index curve from Kittredge and Gervorkiantz (8). Ninety percent of the trees out were in the 35 to 42 year age class. The average age was 38 years. The bolts in sample A, and the corresponding ten- inch test sections, were selected to include bolt diameter classes from three inches to nine inches and bolt height classes from one foot to thirty-three feet. Figure 1 illustrates the distribution of sample sections. The diam- eter class into which the bolts were placed was determined by the diameter inside bark at the lower end. The height in feet of the lower end of the bolt above the ground was assigned as the height class for that bolt. Seven bolt diameter classes were recognized ranging from three to nine inches. Nine height classes were set up at four-foot intervals. The bolts were cut so that the lower ends were either at a height of one foot, five feet, nine feet, etc., up to thirty-three feet when in the stand- ing tree. The bolts were selected so that there was one bolt of each diameter class in each height class in the final sample. For instance, in the five-foot height class there was one bolt for each diameter class making a total of seven bolts. In the selection of sample B, no stratification was made involving height in the tree. Six diameter classes were recognized ranging from three inches to eight inches diameter inside bark.“ Five eight-foot bolts were obtained in each diameter class. The average age of trees in sample B was 33 years. 10 The average height of the ll trees cut to obtain the sample was 47 feet. The site index for this sample was estimated as 65, Just slightly less than for sample A. Determination 92 Weight and Volume Losses Due 39 Drying Weight and volume measurements were taken on both samples A and B every two weeks. Sample A was under study during the summer and fall of 1956, and sample B during the summer of 1957. The four-foot bolts were stacked on a rack construct- ed to afford maximum drying conditions. A rough board roof, supported several feet above the pile, kept off direct pre- cipitation; and the drying rack was built six inches above the ground to allow air to circulate beneath the pile. Since the bolts were protected from direct precipi- tation but were Open to air circulation, the drying condi- tions were severe compared to those ordinarily found in the woods or in concentration yards. The results of any study of air-drying rates are influenced greatly by weather con- ditions and piling methods. For the purposes of this study it was felt that information on maximum drying conditions would be the most helpful. With maximum drying rates approximately established, weight and volume losses in the woods can be assumed to be somewhat less for the same period of year. The bolts were weighed separately, and the weights 11 were recorded by bolt number. The volume of the bolts was determined by immersion in a tank. The inside diameter of this displacement tank was 12.13 inches and the height was approximately eight feet. A sight glass was fitted to the side of the tank; and a steel tape, graduated in inches, was fastened to the glass. Readings were taken before and after a bolt was placed in the tank. The volume equivalent to one inch of displacement in this tank was 0.066 cubic feet. Determinations Made 1 _hg Laboratory in the laboratory a one-inch cross sectional disk was cut from each ten-inch test section, and was used to determine data on the physical prOperties. From these green disks the following data were obtained in the order listed: (1) average diameter both inside and outside the bark, (2) weight of wood, (3) weight of bark, (4) volume of wood, (5) volume of bark, (6) average diameter of heart- wood. The volumes of wood and of bark were determined separately by the water displacement technique. It was found that the bark was easily removed as a continuous ring and could, therefore, be easily immersed. All sample disks of wood and strips of bark were oven-dried. The weights and volumes were again obtained. 'Melted paraffin was used to seal the dried wood and dried bark samples before immersion in water. 12 From the data obtained, the following calculations were made for each disk: (1) moisture content of wood, bark, and both combined, (2) specific gravity of wood on a green and on a dry volume basis, (3) specific gravity of bark on a green volume basis, (4) volumetric shrinkage of wood and of bark, (5) percent of bark based upon diSplacement and diameter measurement methods, (6) percent of heartwocd. These results were averaged by diameter and height classes. A two-way analysis of variance was computed for each of the above mentioned variables, using diameter of bolts and height in the tree as the independent variables. Possible variation of the specific gravity of aspen grown in various stands and sites was thought to be of particular importance. Changes in specific gravity frOm stand to stand would cause corresponding changes in the factor needed to convert green volume to dry weight. In order to obtain some indication as to whether such vari- ations exist, Specific gravity was studied in somewhat greater detail than the other variables mentioned in the preceding paragraph. In addition to the independent variables diameter and height as used in the analysis of variance, a third variable- rate of growth - is known to have an important relation to the specific gravity. A multiple correlation (11) using these four variables was computed. 13 Computation 9: Conversion Factors Conversion factors, i.e., the relationships between green weight or volume and dry weight or volume, were com- puted in two different ways. The first method used was a technique which would be relatively easy for anyone to use who wished to establish a conversion factor for another species. Possible variations and the reasons for them can- not be analysed however. The total green and dry weights and the total green and dry volumes of the test sections were used to establish the desired relationships referred to as conversion factors. The second method of computing the factors was by ' using the moisture content, specific gravity, shrinkage coefficient, and percent of bark by volume us determined from the test sections. See Table III for the formulas used. The emphasis in this study was on this method of computation because it was thus possible to better analyse the various elements which make up the conversion factors. All tables and graphs in the study refer to conversion factors determined in this way. A two-way analysis of variance was computed for each different type of conversion factor using bolt diameter and height as the independent variables, the same as was done when testing the physical prOperties. Three types of conversion factors were computed: (1) those required to convert green volume to dry welsht, l4 (2) those needed to convert green volume to dry volume, (3) those essential for converting green weight to dry weight. The first two of these were computed for both peeled and unpeeled boltwood. In actual practice the conversion factors estab- lished in this study can be used to estimate the weight or volume of oven-dried wood present in aSpen boltwood. The weight or volume of wood at other maisture contents might be desired. This would ordinarily be determined by the moisture content of the product being manufactured. See Appendix B for a discussion of the computation of conversion factors at any moisture content in the case where the second method of computing the factors is used. 'JJ7'7 '1'11! 15 .0009 0500 coca 00 080000 020 Mo 0000000 0 00 cwmmonmxw c .mss0op 00000 000 00 0000000 0 00 00000000m0 .000003 .0000000 005000020 00 00 00 00 00 0 0 00 000000000 00 0000000 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.000 0000 00 0000000 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 00000 00 000000000 .000 0.0 0.00 0.00 0.0 0.00 0.0 0.0 0.00 00000 00 000000000 .000 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00000 .000 .000 -0000 .00 .00 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00000 .0000.000 -0000 .00 .00 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00000 .000 000 -0000 .00 .00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0000 0000 0000 .0.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0000 00000 .0.0 0.00 0.00 0.00 0.000 0.00 0.000 0.00 0.00 0000 00000 0.0.0 0000 .00 0 .00 0 .00 0 .00 0 .00 0 .00 0 .00 0 0000000000 00000. mmwmmHo 00008000,0Hom, «HmowmmnP 0000000 00000000 .00 00000000 .0 000000 00 20000 “0000000 00 0000n00000 00000000 H wqm¢8 16 .0000 00 000000 000 00000 0000 000 00 0000000 .0009 0000 000: 00 .000 000 00 0000000 0 00 commonmxmu .000 00000 00 0000000 0 00 cmmmonmxmo .xhwmv .000000 .00039 .0000000 000000020 ma $.ma 0.mm 0 00 00 00 00. 00 00 00 00 0.00 0.00 0.00 0.00 0.00, 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.000 0.00 0.00 0.00 0.00 _ 000300005 no pcmonom H.Ho> thm 00 pcmohom 0&Hom Ho mwwxcwnnm .Hob 0.0 0.0 0.0 0.0 0.0 0.00 0.0 0.00 0.0 0.0 00000 00 000000000 .000 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00000 .0o>.000 -00 .00.00 00.0 00.0 00.0 00.0 00.0 00.0 “00.0 00.0 00.0 00.0 0000m .0o>0.000 u; .0u.mm 00.0 00.0 00.0 00.0 00.0 00.0 .00.0 00.0 00.0 00.0 00000 .000 000 -00 .00.00 0.00 0.00 0.00 0.00 0.00 0.000 0.00 0.00 0.00 0.00 0000 0000 0000 .0.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0000 00000 .0.0 0.00 0.00 0.00 0.00 0.00 0.000 0.00 0.00 0.00 0.00 0000 00000 0.0.: 0000 00 00 00 00 00 00 0 0 0 0000000000 00000 0000000 mpnw0wm 00000000 mmmmdqo BmOHflm Mm mmwdmm>d .0 000000 00 20000 00000000 00 0000000000 00000000 HH fiqmda 17 TABLE III DEFINITION OF COFVFRSION FACTORS, AND THE FORMULAS USFD TO CONIUTE CONVERSION FACTORS Sygbol Explanation CF-l ....Factor by which to multiply the green vol. of wood plus bark, in cu. ft., to obtain the dry weight of wood, in pounds. CF-2 ....Factor by which to multiply the freen vol. of wood only, in cu. ft., to obtain the dry weight of wood, in lbs. CF-S ....Factor by which to multiply the green vol. of wood plus bark to obtain the dry vol. of wood. CF-4 ....Factor by which to multiply the green vol. of wood only ‘ to obtain the vol. of dry wood. CF-S ....Factor by which to multiply the green weight of wood plus bark to obtain the dry weight of wood. B .......Ratio of the green bark vol. to vol. of wood plus bark. 8 .......Ratio of the vol. of wood shrinkage to vol. of green wood. SGwd ....Specific gravity of wood on a dry vol. basis. SG g ....Specific gravity of wood on a green vol. basis. SGbg ....Specific gravity of bark on a green vol. basis. MCw .....Moisture content of green wood. dCb .....Moisture content of green bark. Formulas Used to Compute Conversion Factors CF-l = (1-13) (l-S)(62.32) (SGwd) ‘ CF-z (1-S)(62.32)(SGWd) (BF-3 = (l-BHl-S) CF-4 = 1-8 CF-S 18 .mHOpowm noamnm>coo no coflpmcoamxm Hey HHH canoe comm mm¢.o bm¢.0 mm¢.0 00¢.0 0n¢.0 ¢H¢.0 mm¢.0 www.0 mime nom.0 000.0 mum.0 H00.0 000.0 0H®.0 mam.0 000.0 ¢Ih0 mmb.0 ¢b5.0 mmb.0 $55.0 omb.0 dou.0 m05.0 _ n5>.0 nufiv Ho.nm 0H.mm ww.¢m 0¢.¢N 00.nm ww.mm om.mm b¢.mm Nlho 0*.0m H5.HN m¢.HN Hm.0m o¢.0N ob.ma n¢.ma bw.mH. Aime Gama .GH 0 .nH 0 .GH 5 .GH 0 .GH 0 .GH fl .GH 0 amHOHOwh cacao mommoao HmamSaHQ paom cofimnmpnoo mammaqo mmemsaHm am mmoamm>e .4 mnmzem 20mm mmapmaoo zmmma meooeaHm mom mmoeoem aonmm>zoo >H HamwB 19 .poou :0 undonw map mpopm paon on» we pnwfiom p .mnouomm nofimnw>coo ho qofipmqoaaxo How HHH manna 0000 000.0 500.0 0av.0 0H¢.0 000.0 HH¢.0 500.0 000.0 “000.0 000.0 M mumo 000.0 ¢00.0 0H0. 000.0 000.0 000.0 H00.0 001.0 H$00.0 000.0 filmo 005.0 055.0 055.0 005.0 H05.0 055.0 005.0 005.0 505.0 005.0 nuho H0.00 $0.00 00.00 00.00 50.00 m .00 00.00 00.00 00.00 00.00l 0-00 00.00 00.0H 00.00 00.00 00.H0 um.na 00.00 00.00 00.a0 0¢.HOW Humo cams an mm mm Hm ea ma 0 m H {w emanpoma compo mommmau ppnmfiwm nonH¢>noo mmmmaao amonm am amoamm>< .4 matuam Acme QEBDaéOU ZflAmd EBOOBQHm mom > wqm¢B mmohomm .mmHmmM>300 III. RESULTS AND ANALYSIb Conversion Factors Based Upon Actugl Total Weights and Volumes The conversion factors desired were first computed from the total weights and volumes of the test disks. For example, the factor used to convert green volume in cubic feet to dry weight in pounds was found by dividing the total number of pounds of dry wood in the one-inch thick test disks by the number of cubic feet of green wood and bark in these disks. Using this method of calculation the conversion factor for aspen used to convert the green volume of wood plus bark to the dry weight of wood was found to be 21.06. Thus, a load of unpeeled boltwood with a volume of 1000 cubic feet, as determined by immersion, contains an esti- mated 21,060 pounds of dry wood substance. The six following conversion factors were determined from the total weights and volumes of the test disks. To estimate the dry weight of wood in pounds: (1) Multiply the green volume of unpeeled aSpen, in cubic feet, by 21.06. H) (2) Multiply the green volume 0 peeled aspen, in cubic feet, by 24.30. (3) Multiply the green weight of unpeeled aspen 2O 21 by 0.422. (4) Multiply the green weight of peeled aspen by 0.507. To estimate the dry volume of wood: (5) Multiply the green volume of unpeeled aspen by 0.781. (6) Multiply the green volume of peeled aSpen by 0.901. It will be seen that the first two conversion factors vary approximately 3 percent from the some factors as calculated by the second method, which will be discussed later. The differences are believed to have occurred as a result of errors in the measurement of the dry volume of the test disks. These measurements were used only in com~ puting the factors by the second method. The remaining portions of this paper deal with the physical prOperties of aspen and with conversion factors derived and analysed in a different way and in more detail than was done in the preceding discussion. Physical Properties 2; giggooth Aspeg The determination of average conversion factors and of the variation of these factors were two of the primary objectives of this study. The significance of differences in the conversion factors between diameter and height classes was established by means of analysis of variance. The conversion factors, however, were computed from a number of variables such as specific gravity, percent of TABLE VI SUHLAHL OF THE RESULTS OF ANALYSIS OF VARIANCE ON PHYSICAL PROPERTIES OF BIGTOOTH ASPEN variation Variation Between Between Physical Sample Bolt Diem. Height Properties Mean Classes Classes M.c.a Green wood 93.9% NS NS L3G. Green Bark 71.9% NS ** Sp. Gr. wood on Dry 0.42 ** ms Vol. Basis Sp. Cr. Bark on 0.60 NS NS Green Vol. Basis Vol. Shrinkage of 9.79 NS NS Wood Bark Volumeb 15.4% NS * NS.... Analysis of variance showed no significant differences between the means of the classes. *..... An actual difference between the means of the classes was indicated at the 5% level. **.... An actual difference was indicated at the 1% level. 8Moisture content. b bark. Expressed as a percent of the vol. of wood plus 23 mom.o u pamaofloomou qofipmamnnoo onHpHsu ono.o H opwswpmu on» no nonnm onmocmpm naaoo.on mg¢oo.o- Assoc.o+ moe.o n w I\ m.X CO V MO smm.n ann.u mma.+ mcoammmnwoz onwoqmpm m.N SPHB V ¢¢H.s mon.u Hb¢.+ Mo mcofipoawnnoo as me Hm w coo: no mmdonu obop< nwm nocH nmw.mwcfim mpHom mo .Eoma mafipmno ommHoomm momaHma> azmazmmmo was goes so waHpamo onHommm .zQHmmmmumx mquagpu may no menpmmm HH> Mdmda 24 .ahop 03mg venomous hopes—mg :on on goon oasaob be a no cool no hpggw 033on Mo nowpoaon on» yo ”30.5 .N .w: mm102_ l EMFMESQ ._.I_Om m N m n v m wm. snsve "10A ABC! All/“189 OldiOI—st bark, and percent of shrinkage of wood. The extent of change of these variables will be considered first in order that the variations found in the conversion factors may be better understood. The Specific gravity of wood is an important ele- ment used in the formula for converting green volumeto dry weight, and for this reason it was investigated with the aid of both analysis of variance and multiple corre- lation. See Table V1 for the results of analysis of variance and Table VII for the results of the multiple correlation. Specific gravity based upon dry volume rather than green volume was used in the analysis of specific gravity in this work. This was done because weight per unit of dry volume was needed when computing the conversion factors. Specific gravity was found to vary significantly between bolt diameter classes, but not between height classes. Figure 2 illustrates the increase of Specific gravity with diameter. The increase of specific gravity with diameter could be partially due to differences in both the amount of heartwood and the growth rate. The percent of heartwood was somewhat more constant beween diameter classes than was the rate of growth. Heartwood varied only from 9 to 18 percent. Since the rate of growth varied from 6 rings per inch to 22 rings per inch this variable was investigated further. The correlation 26 coefficient of bolt diameter and specific gravity was found to be slightly higher and Opposite in sign than the corre- lation coefficient of rings per inch and specific gravity. Refer to Table VII. The correlation indicated that the specific gravity increased with an increase in diameter, but decreased with an increase in the number of rings per inch. The standard error of the estimate for the multiple regression was found to be 0.036. This indicates that the regression formula does an adequate Job of estimating the Specific gravity. The multiple correlation coefficient of 0.506 was significant at the one percent level. When considering the partial regression coefficient of rings per inch on specific gravity it is to be remem- bered that if ring width were used instead of rings per inch the regression coefficient would be positive in sign instead of negative. Paul (10) states that work at the United States Forest Products Laboratory on five Species of the genus Populus indicates that there is a negative regression for ring width on specific gravity. The reason for this contradiction cannot be eXplained by the writer. The average volumetric shrinkage of wood based upon green volume wasfound to be 9.7 percent.' This figure did not vary significantly with either bolt diameter or height classes. The percent of bark volume is the third important 27 .Mnop mean coo: no oadaop on» no unconog o no commonmuo on ”non no oesaop .oona on» an vawnon nun: anon macaw no ossaop one :« nonaonnep on» no ensue .n .mnh .._..._ l ._.Io_m_I 0? mm VN m. m 0 E o N MHVB :JO .LNEOHBd 28 .oona on» an Emacs up: anon noonw no unopaoo 83305 no uoupoung one no nacho .d .3.“ ....n. .. PIGEI as an em o. m __ _ n l u , . _ 0 LO LO _ <1. £0 NN Om mm iwaoaad — lwaiwoo Banismw 29 TABLE VIII SUMMARY OF THE RESULTS OF ANALYSIS OF VARIANCE ON CONVERSION FACTORS variation Variation Between Between Conversion Sample Bolt Diem. Height Factorsa Jean Classes Classes CF—l 20.46 ** we CF—Z 23.61 * NS Ubbfi 0.782 NS NS CF—5 0.425 NS NS NS... Analysis of variance showed no significant difference between the means of the classes. *.... An actual difference between the means of the classes was indicated at the 5% level. **....An actual difference was indicated at the 1% level. 6See Table III for an explanation of conversion factors. 30 m .kovoadnc vaon new. ado Henge Hoaoon nonaohnoo no .3332: on» no "395 .m .mrn mum—.52. l ¥m o>fiwmassso pmmoa $m.am $0.0m ma.ma &H.ma ab.ma mm.oH m¢.b mm.¢ --- pnmfima mpfipmHsSSU xnmm mow wmm fimm fiao ~ moo mop flab mmm Rom msam coo: “0 .o.m m A w .»m mfl.mm mm.bm po.mm_ on.mm. m¢.mmm wo.mm um.wm 0 .mm mm.mm .so cH .Ho> Hmpoe nowm mflwm momm mmmm muom~ owbm mmmm Hnmm coon w.mnq a“ .pz Haves ona pm mm mm mm m« mm «a mo < wagsmm unmSmnsmmwz HwfipficH mouflm cmmmwaw mama mo .0: mZOHBHQZoo OZHWchmHfi QOOO mHQHD Dfl>mmmmo .4 anflfim SOHh mpflmfimhdmwwu .meqom Ewmmd QMAMMMZD Boomlmpom ho mammoq MEDHO> nfid Emont %H mammB .pqumHSmwwS cmopm Hmflpfisfl wgp ho unwopmg m mm owmmmhmxm n .bmma .HN wand no :mxwp mums mwcfiomwn HprHCHm flow.a me.H sam.o --- gmmog ¢23H0> w>wpsaxs m mm.¢a éw.o fin.g In: pmmoq pgufiwe ¢>Hpmasuso wbo web QMt “MU xgwm mafia coo: go .onfl o¢.am oo.am mu.am Hm.am .pm .30 zH .Ho> Hmpoe may wuw maoa mboa .mp4 CH .p? ampOH aw pm ma. Db m wamemm pawsmpsmmm: HmfipHcH mocfim ommamam mmwm mo .02 *4 Gong mpstwpswwmu 35 .Mqum,QmAMMHHD anb.m Rum.a flow. nun pmmoa madao> o>Hpmassdo mm.om ,o.wm venom anJ pmmod anwmc mpwpma5850 mow wow aha mam xnmm mafia coo: mo .o.x om.ma om.ma . wm.wa _ mo.ma .pm .50 a“ .Ho> proe mmo «no Heb mum .mna aw .pz Hmpoe HW hm, NH LDo m,maa8mm psoSmndmmmu HmfipfinH wonam cmmmwam mqu mo .02 *4, somh,mpn08wnsmmwm msgom,omqmmm mZOHBHQZOO wszmQIde Doom meZD Qw>mmmmo .m EAAH Q24 BEQHWS N mqmOZ _ .HOO.‘-.._.n_mm — .034 _ ‘ _ . r... ON 0? 0m .._.>> Om "IIOIflO/ IO/OI/O/nvl 9 ...O> . .. OO. awn'lOA cmv J.H9|3M .‘Ivmslao JO ‘iNaoaad 37 .m cannon Bonk ohm «nun .nqodaaonco waahncauad coom Mona: undon.aonuu anaconda can voaoom anon nu ON_ 8:230 333 25:3 can £32. 3:65 £86 .m .wE F30 m02_m m>43... _ I 938 033sz no 5.6.2., I mfiom Pflmmaz: ...o m230> om oia mica omjmma ...o :66; . 9..-; whom omqmma ...o m230> O? I Om / 7;, fiom OO_ BWFHOA ONV .LH9I3M' ‘IVNISIHO :JO iNBOHBd .noapuooa comomxo an a“ wnahncuuau manage upaop comma anaconda can ooaoom no advance ondpmuqs one we nacho .o .wgh ...Do moz_m w> maoa. am ®¢OH0HH Cowpowh OD nmam.HH New » eeH.- m.“ ama.am mma.mm .opw .«mm> ama.nn eoo.eamo .osm am aea.ama oan.eanma seapomnmoo n omw.mee ooo.emmmm .ops .Nxm a manol Ownol m.H one.a ooa.smnm oas.mm .opm .«am> noa.n- Hmo.enm- amm.wmm .opw am oma.eam Hoa.aaaaa no .mnma acapomn co m aam.aam ooa.moHHH own.eeom .opm .mxm a Haw. mao.o emm.u . m.n eao.m oan.nmma www.mae aaa.mH .opw «ama oan.m aoa.aa- onn.omn- oam.anm .opm m Heb.mma aoa.amwo oaa.oame oom.aomm conpomnmoo H HnH.H©H oon.mneo omn.onmn oaa.womm .opm .mxm a was. ao.aa ma.ma 00.9 cmma m¢.om maoH o.mos H.wan saw » ma ma HM mammm .Ho> mam no ccdoac xnmm ooHaQH coon no .no .am mpopa paaamaw zoaH pom mmcwm paom mo .amwa Zfiwmd SBCQBOHm ho ozofid mBZmHUHthOQ mOHBmAMmmoo mo HM Mazda mEZHzmmDmmmu mach ”CHemBDwgoo TABLE XII ANALYSIS OF VARIAYCE OF SIECIFIC GRAVITY OF WOOD ON A DRY VOLUME BASIS Source Cir“ Sum 0? Degrees Yeah Variation Squares of Square F F r F Freedom '00 '01 Diameter .0274 6 .0045? 5.44“ 2.30 3.20 Height .0153 8 .00191 1.44 2.14 2.90 Error .0638 48 .00133 Total .1065 62 ANALYSIS OF VARIANCE 0F TIE MOISTURE CONTENT OF GREEN HEARTWOOD AND SAPWOOD COKBINED Source of Sum 5?: Degrees Mean variation Squares of Square F F.05 F001 Freedom Diameter 2783.3 6 463.80 2.10 2.30 3.20 Error 10627.8 48 221.41 Total 15132.7 62 ANALYSIS OF VARIANCE OF THE PERCENT OF‘WOOD TABLE XIII SHRINKAGE TO AN OVEN-DRIED CONDITION Source 6?6 Sum of' egress Mean Variation Squares of Square F F.05 F001 Freedom Diameter 32.94 6 5.49 2.21 2.30 3.20 Height 8.21 8 1.03 .42 2.14 2.90 Error 119.02 48 2.48 Total 160.17 62 ANALYSIS OF VARIANCE OF THE RATIO GREEN. BARK VOLUHE OVER VOLUKE OF WOOD AND BARK Source of Sum 5?' Degrees Yean‘ Variation Squares of Square F F F .05 .01 Freedom Diameter 26.80 6 4.47 .87 2.30 3.20 Error 247.16 48 5.15 Total 380.50 62 *Indicates significance at the 5% level. TABLE XIV ARILYSIS OF VARIANCE OF THE MOISTURE COI‘ITFIIT OF GRIT-71‘;- BARK Source of Sum of* Degrees Mean Variation Squares of Square F F 5 F 1 Freedom '0 ‘0 Diameter 285.91 6 47.63 .98 2.30 3.20 Height 2394.61 8 299.33 3.13” 2.14 2.90 Error 2343.25 48 48.82 Total 5023.77 62 nNALYSIS CF VARIAHCE OF CONVERSION FACTOR ITLL'BER mama Source of Sum of Degrees Yean Variation Squares of Square F F 05 F 01 Freedom ‘ ’ Diameter .005 6 .0008 1.33 2.30 3.20 Feiflut .00? 8 .0010 1.67 2.14 2.90 Error .030 48 .0006 Total .043 62 8See Table III for explanation of conversion factor. *‘Indicates significance at the IA level. TABLE XV AFALYS'S OF VARIAECE 0F CELVERSIUN FACTOR IL.’ . E91 CITE Source of Sun of Derrees Pean Variation Squares of Square F F F .05 .01 _ Freedom Diameter 47 6 7.80 5.7g** 3.53 3.30 Heirht :23 8 2.30 1.41 2.14 2.90 Error 91 4? 2.06 Total 69 62 23.11.23 a CF Vinins‘cr 03‘ ‘35 VIZ-(SIZE? FACIESIIIH.BLR FOUR Source of Sum of Derrees lean Variation Squares of Square F F F .05 .01 Freedom DIfiinE-‘LEI' oJQU‘fl 6 oC‘JOC? 2.25 2.30 3.20 Bright .080? 8 .00011 .44 2.14 2.60 Trror .0118 4? .52025 Total .0161 62 8See Table III for explanation of conversion factors. **Indicates significance at the 13 level. (I! TABLE XVI AKALVSIS OF VARIATCF CF gQVVERSICY FACTS} EUiBER TWC Source of“ sum of * earees Tean Variation Squares of Square F F 05 F 01 Freedom '“U ' Diameter 67.24 3 11.207 3.00* 2.30 3.20 Error 179.19 48 3.733 Total 288.53 62 - 13' w AIRLYbI 3 OF VARIAECE 02 001vr3310r F‘CTOR NUMBER FIVEa Source 6?’ ‘Sum of: Degrees Teen Variation Squares of Square F F 05 F F1 Freedom ’ '“ Diameter .3060 6 .0010 1.05 2.30 3.20 Weight .0112 8 .0014 1.47 2.14 2.90 Error .0459 48 .000?5 Total ' .0631 62 8See Table III for explanation of conversion factors. *Indicates sirnificance at the 5% level. APPENDIX B. DETERMINATION OF CONVERSION FACTORS USED TO ESTIMATE THE WEIGHT AND VOLUME sT ANY MOISTURE CONTENT The conversion factors presented in this work can be changed in such a way as to be used to estimate the amount of wood, by weight and volume, present at moisture contents other than zero percent. In the formulas used to compute factors for conversion to dry volume, i.e., factor number three and number four, only the‘percent of shrinkage portion needs to be changed from that shown in the formulas in Table III. The percent of shrinkage to zero percent mois- ture content is multiplied by 2938;! to give the percent of shrinkage at the moisture content in question. M re- presents this moisture content. When computing factors to convert green volume to the weight at moisture contents other than zero percent, the shrinkage portion and the specific-gravity portion of the formulas for factors number one and number two must be changed. The factor must then be multiplied by one plus the moisture content to allow for the weight of the addi- tional water. The Specific gravity on a basis of volume at zero percent moisture content can be changed to Specific gravity 56 57 at other moisture contents by use of the formula; where Sd represents the specific Sd Smas— l + T. 009 WSdWH) gravity on a basis of volume at zero percent moisture content, and Sm represents the specific gravity on a basis of the volume at the desired moisture content. This formula is eXplained by Brown, Panshin, and Forsaith (l). '1) f n') ~L 10. ll. BIBLIOGRAIHY Brown, H. F., Ianshin, A. J., and Forsaith, C. C. Textbook of Wood Technology. Vol. II, New York: NcGraW-Hill Book Company, 1952. Chapman, Herman H., and Demeritt, Dwight E. Elements of Forest Iensuration. Albany: hilliams 1ress, 1036. Goulden, Cyril H. Kethods of statistical Analysis. New York: Joh Niley and do: , 1952. Hale, J. D. ”Thickness and Density of Bark for Six Pulpwood Species", Iulp and taper Kogazine of Canada, December, 1055. James, Lee T. marketing lulfwood in Xichigan. Agricultural Piperiment Station, Lichigan State University, Special Bul. 411, February, 1957. Jeffords, A. 1., Jr. "Trends in fine Iulpwood marketing in the South", Journal of Forestry, September, 1956. Jensen, Raymond s., and Davis, John R. Seasonal Toisture Variations in ASIFH. Jinn. Forestry Notes, No. 19 University of Yinr., 3t. Iaul, July 15, 1955. Kittredee, Joseph, Jr., and Gervorkiantz, S. R. Forest Iossibilities of Aspen in the Lake States. Hinn. Agricultural Fxperia nt Station, Tech. Bul. 6C, 192?. » Olson, Harold S. The Iurchase of wood by Weight, lroceedings of the typer kississirpi Section of the Forest Lroducts Research dociety, October 6, 1655. laul, Benson H. Specific Gravity of loyulus species and Hybrids. nited states Forest lroducts Laboratory, Vadison, Report No. 2060, August, 1956. Snedecor, George W. Statistical Yethods. Ames: Iowa State College Iress, 1046. ‘ 1‘]! ll'.’ 11.1..I1}; : ‘1‘. ‘11., Dill .IVuaui-I.|.II! .1 ‘ A l2. Taras, Kichael n. Buying lulpwood by Weight. Southeast Forest Experiment Station, Asheville, Station Taper No. 74, November, 125”. 15. Yandle, David 0. Statistical Evaluation of the Effect of Age on Slecific Gravity in Loblolly Tine. United States Forest lroducts Laboratory, Madison, Report Do. 2049, February, 1956. Demco-293 ——_—_________,,._..-——- ”11111111111?“ [111311111 1111116113111“