ABSTRACT Eight coniferous and two deciduous tree species common to Wisconsin were evaluated for wood density. Overall core and tree specific gravity means for species are presented, as well as means and variation by diameter class within Forest Survey Units. The northeastern survey unit of the State had significantly higher core specific gravity for six species. For four other species the values for this area were higher but not significantly. No other trends were found.

ACKNOWLEDGMENT The assistance of personnel from the Wisconsin Department of Natural Resources, Nekoosa-Northern Paper Company, the Chequamegon and Nicolet National Forests, and the North Central Forest Experiment Station made this study possible. Their cooperative assistance is gratefully acknowledged. Special thanks to Frank Freese of the Forest Products Laboratory for assistance with statistical design and analysis.

-i-

WISCONSIN WOOD DENSITY SURVEY By Robert R. Maeglin, Forest Products Technologist Forest Products Laboratory1.

Forest Service

United States Department of Agriculture

INTRODUCTION Since 1928 the U.S. Forest Service has been engaged in a continuous inventory of the Nation's timber resource. This inventory is accomplished by field surveys and provides for an appraisal of the area of forest land, volume of timber, growth, growing stock estimates, species composition, and other resource statistics. Not until 1956, however, was any consideration given to an evaluation of the intrinsic quality of the wood produced by the forests. In 1956, coincident with the resurvey of forest resources in the State of Mississippi, a wood density survey was begun. That survey was the beginning of a nationwide program of wood quality assessment, using wood density as the criterion for evaluation. Wood density is a measure of the component structure of wood and is directly related to strength and stiffness of structural lumber as well as to pulp yield. Wood density also directly affects the density and properties of particleboards and other reconstituted panel material. Reports on the wood density for more than 25 softwood and soft-hardwood species from the West, South, and Northeast United States have been published (5-8, 22, 29, 32, 33, 36, 37, 39, 41, 42).2 This report provides similar data for 10 species in Wisconsin. The resurvey of forest resources in Wisconsin was begun in 1967. This provided the opportunity to conduct a wood density survey in conjunction with the Forest Survey. The wood density survey was conducted jointly by the U.S. Forest Service's North Central Forest Experiment Station (NC) and Forest Products Laboratory (FPL), and the Wisconsin Department of Natural Resources (WDNR). The species evaluated were: Balsam fir (Abies balsamea (L.) Mill.), tamarack (Larix laricina (Du Roi) K. Koch), white spruce (Picea glauca (Moench) Voss), black spruce (P. mariana (Mill.) B.S.P.), jack pine (Pinus banksiana Lamb.), red pine (P. resinosa Ait.), white pine (P. strobus L.), eastern hemlock (Tsuga canadensis (L.) Carr), bigtooth aspen (Populus grandidentata Michx.), and quaking aspen (P. tremuloides Michx.).

1Maintained

at Madison,

Wis.,

2Underlined

numbers

parentheses

in

in

cooperation with refer

to

the University of

Literature

Cited

at

end

Wisconsin. of

this

report.

OBJECTIVES As with the other wood density surveys, a series of objectives was established. Wisconsin survey objectives were: to

The

1. To establish the mean and variation of wood density for each species sampled and establish the magnitude of wood density differences between species.

2. To investigate any trends of wood density in relation to environmental or geographical factors. 3. To locate, by systematic sampling, trees of superior quality in growth, form, and wood density for genetic study and propagation. Throughout this paper the terms density and specific gravity are used interchangeably. Where values are given, density is expressed as pounds per cubic foot (p.c.f.). All values of density are based on green volume and ovendry weight. PHASES OF STUDY Phase I Phase I is the systematic collection of increment cores from forest survey plots. This phase was conducted by the North Central Forest Experiment Station and the Wisconsin Department of Natural Resources.3 Details of sampling and the later processing are included in the appendix. All cores collected were labeled and shipped to FPL for processing. Phase II Phase II is concerned with laboratory processing of increment cores and the statistical analysis and interpretation of the data. All work on this phase was conducted by FPL. The primary objective was to establish the mean and variation in wood density (both core and tree values) for each of the species involved. The values calculated from the core specific gravity data included the mean, standard error of the mean, standard deviation, and range for increment core specific gravity. These values combined with information from phase III were then used to calculate the mean, standard error of the mean, and range for estimated tree specific gravity. Formulas are given in the appendix. Phase III Phase III is concerned with the development of regressions to predict merchantable tree specific gravity from increment core specific gravity data collected in phase II. Phase III was conducted by FPL and details are given in the appendix.

3

Robert N. Stone (now of the Southeastern Forest Experiment Station) and Burton L. Essex of the North Central Forest Experiment Station were in charge of forest survey for the U.S. Forest Service, and Stanley W. Welsh was in charge for the Wisconsin Department of Natural Resources. -2FPL 202

Using the general linear model,

regressions were fitted with tree

specific gravity, the dependent variable, and core specific gravity, d.b.h., age, reciprocal of age, and d.b.h. 2 as potential independent variables. Separate regressions were developed for all diameters combined and for 2-inch diameter classes (4.0-5.9, 6.0-7.9, 8.0-9.9, 10.0+) for all species except eastern hemlock, bigtooth aspen, and quaking aspen which had two different classes (10.0-11.9 and 12.0+). RESULTS

AND

DISCUSSION

Tree Specific Gravity Prediction Equations Table 1 lists the equations used to convert the core specific gravity values to tree specific gravity. Selection of the combinations of variables used in the table 1 equations was based 2 on the frequency of occurrence--by R ranking--for the variable combinations. For example, in jack pine the combination of core specific gravity and age is used. This was established by assigning a ranking of 1 for the best combination, 2 for the next best, and so on. Core specific gravity and age had the lowest cumulative total (11) for all five diameter classes. The next lowest combination had a cumulative total of 13. For other species, variable combinations were core specific gravity and d.b,h., and core specific gravity and reciprocal of age. This system of selection was used to obtain a consistent set of variables for all diameter classes within a species. As such, this procedure results in some regressions being nonsignificant at the 0.05 level of probability. Table 2 shows averages and ranges of data from which the equations were developed. Wood Density Evaluation Variation.--Table 3 presents data by diameter classes within survey units (areas shown in fig. 1). Included are means, standard errors, and standard deviations. The standard error (S –) provides an indication of the precision with which X mean estimates the population mean. As a rough guide we can say that unless chance has occurred in sampling, the sample mean will be within two standard of the population mean. Standard errors could not be computed for the means based on samples from less than two locations. The standard deviation is a measure Standard deviations in this report were gravity values, using tabular values of Values for standard deviation were not or five trees in the sample.

the sample a 1-in-20 errors that were

of the variation of individuals about the mean. derived from the sample range of core specific the ratio of standard deviation to the range (11). calculated if there were less than two locations

Another expression of core specific gravity variation is the coefficient of variation (CV) , which allows comparison of the variability about different-sized means (11) . Coefficients of variation are shown in table 3. The lowest relative variability is found in the northwest survey unit with plantation-grown red pine (3.3 pct.). The highest relative variability is found in the northeast survey unit with bigtooth aspen (19.4 pct.). -3-

Trends in core specific gravity.--A comparison of core specific gravity by survey units in table 3 indicates one consistent trend or pattern. That is, where sampling is intensive enough to permit comparison, the core specific gravity values for the "all diameter" class in the northeast unit are generally higher than those for the other survey units. One exception is the central unit for quaking aspen. The differences between units in the "all diameter" class (where five or more samples are involved) within species is quite sizable, the greatest difference being 0.048 between the northeast and northwest units with natural red pine. For balsam fir, black spruce, jack pine, and bigtooth aspen, the greatest difference between units, within the "all diameter" class, is 0.016. For the remaining species the differences are between 0.020 and 0.038. When unpaired "t" tests are made of the greatest differences between units in the "all diameter" class, as noted above, we find the following: Species

Units Compared

Balsam fir Tamarack White spruce Black spruce Jack pine Red pine (natural) Red pine (plantation) White pine Eastern hemlock Bigtooth aspen Quaking aspen

NE and NW NE and Cen NE and NW NE and NW NE and SW NE and Cen NE and Cen NE and Cen NE and NW NE and Cen Cen and SE

Figure 1.--Forest survey units in Wisconsin.

FPL 202

-4-

Significant at P = 0.05 Yes Yes Yes No No Yes No Yes No Yes Yes

There is no apparent explanation for the higher specific gravity values in the northeast unit. However, meteorological records (34) indicate that this unit has the lowest average annual precipitation of the five units. Also, the northeast unit has a low average warm season (May-September) rainfall. Average Annual Forest

Survey

Northeast Northwest Central Southwest Southeast

Unit

Precipitation In. 29.92 30.02 30.64 31.54 30.41

Average

May-September Rainfall In. 17.66 19.10 19.94 19.60 17.09

There are no strong patterns of specific gravity between diameter classes. A great deal of uniformity is shown in some species, for example, balsam fir; the northeast unit has a total specific gravity difference of 0.008 between diameter classes and the northwest unit a difference of 0.017. Bigtooth aspen is a little less uniform, displaying differences of 0.019 for northeast, 0.016 for northwest, 0.021 for central, and 0.032 for the southwest unit. Quaking aspen is even less uniform, with differences of 0.019, 0.015, 0.015, 0.057, and 0.027, respectively, for the same order of units. The greater differences shown in these and for other species are primarily due to small sample size. Table 4 presents data on species ranges and averages for the mass sampled trees in this survey. For comparison, table 5 presents data from other sources, including ranges or means, and locations from which the material came. While some variation exists in both ranges and means between tables 4 and 5, most species values are in reasonable agreement. The divergence in values can result from several sources including genetic variation, sampling variation, and age differences. For example, average jack pine core or sample specific gravity in tables 4 and 5 varies from 0.40 to 0.42 except for Maeglin's (21) data with a specific gravity average of 0.337. The average age of the trees having low specific gravity is one-half or less that of the other sources. Wisconsin-Maine comparison.--A comparison, using unpaired "t" tests, between species common to both this study and the Maine Wood Density Survey (39) indicates the following: White spruce, black spruce, white pine, and eastern hemlock core specific gravities are not significantly different; balsam fir, tamarack, and natural red pine core specific gravities are significantly different at the 0.05 level of probability, Trends in tree specific gravity.--The higher value noted for the northeast unit core specific gravity still generally holds for the "all diameter" class. But the value of bigtooth aspen from the southwest unit and the value for quaking aspen from the central unit exceed those of the northeast unit. In table 3, tree specific gravity differences between units within diameter classes are generally lower than those for core specific gravity. An exception is found with white spruce where differences increased in the 6.0 and 10+ inch classes. Other than the generally greater average "all diameter" specific gravity of the northeast unit, there were no geographic trends. Figures 2A-12A in the appendix give core specific gravity, d.b.h., and age information by counties. A warning must be given, however, that these are but estimates and generally based on a small sample. These data are presented only as general information as to the form of the sample. No evaluation of height-specific gravity relationships were considered, as they can be found in individual reports (1, 20, 25, 27). -5-

SUMMARY AND CONCLUSION The Wisconsin Wood Density Survey, conducted in conjunction with the periodic forest inventory of the State of Wisconsin, provides information on the basic wood density of the principal softwood and soft-hardwood species of the State. Species evaluated were balsam fir, tamarack, white spruce, black spruce, jack pine, red pine, white pine, eastern hemlock, bigtooth aspen, and quaking aspen. Data presented include ranges of core and tree specific gravity, d.b.h., and age, and means for each species as a whole; core and tree specific gravity means, standard errors, and standard deviations by diameter class within forest survey units and species; and coefficients of variation for core specific gravity by diameter class within units and species. A tabulation of specific gravity and related data from other studies is presented for comparison by species. For all species except quaking aspen, core specific gravity was higher for the northeast survey unit than for other units, where the sample size was five or greater. Unpaired "t" tests between units indicate balsam fir, tamarack, white spruce, natural red pine, white pine, and bigtooth aspen had significantly (P = 0.05) higher core specific gravity in the northeast unit. Black spruce, jack pine, plantation red pine, and eastern hemlock were not significantly different (P = 0.05) between units. The difference in core specific gravity between units in quaking aspen was significant (P = 0.05). A "t" test of species common to both Wisconsin and Maine indicates no difference between states in the mean specific gravity of white spruce, black spruce, white pine, and eastern hemlock. A significant (P = 0.05) difference was noted for balsam fir, tamarack, and natural red pine. These data will provide the bases for determining weight yields, pulp yields, and estimates of strength. Other applications would include weight estimates for balloon and helicopter logging.

FPL 202

-6-

Table 1.--Regression equations for predicting tree specific gravity from

increment core specific gravity and related factors, for

species by diameter class

(Page 1 of 2) -7-

Table 1.--Regression

equations for predicting tree specific gravity from

increment core specific gravity and related factors, for

species by diameter class--continued

1C = core specific gravity A = age D = d.b.h. R =1/age ** Significant at the 0.01 level of probability. * Significant at the 0.05 level of probability. NS Not significant at the 0.05 level of probability. FPL 202

-8­

(Page 2 of 2)

1

Values for plantation red pine include 8 trees less than 4.0 in. diameter and 3 trees greater than 9.9 in. diameter that were used in developing the “all diameter” equation but not used in the diameter class equations.

Table 2.--Rangesand averages of data used in phase III regression development

Table 3.--Specific gravity and related data by diameter class

within species and forest survey units

(Page 1 of 7)

FPL 202

-10-

Table 3.--Specific gravity and related data by diameter class

within species and forest survey units--continued

-11-

(Page 2 of 7)

Table 3.--Specific gravity and related data by diameter class

within species and forest survey units--continued

(Page 3 of 7) FPL 202

-12-

Table 3.--Specific gravity and related data by diameter class within species and forest survey units--continued

(Page 4 of 7) -13-

Table 3.--Specific gravity and related data by diameter class within species and forest survey units--continued

WHITE PINE--continued

EASTERN HEMLOCK

(Page 5 of 7) FPL 202

-14-

Table 3.--Specific gravity and related data by diameter class within species and forest survey units--continued

BIGTOOTH ASPEN

(Page 6 of 7) -15-

T a b l e 3 . - - S p e c i f i c g r a v i t y and r e l a t e d d a t a by d i a m e t e r c l a s s w i t h i n s p e c i e s and f o r e s t s u r v e y u n i t s - - c o n t i n u e d

QUAKING ASPEN

1 2

CV = coefficient of variation.

Spencer, J. S., Jr., and Thorne, H. W. Wisconsin's 1968 forest resources. Res. Bull. NC-15. NC Forest Exp. Sta., St. Paul, Minn. 1972. 3 Insufficient data to develop a prediction equation.

U.S. Forest Serv.

(Page 7 of 7) FPL 202

-16-

Table 4.--Specific gravity and related data for 10 Wisconsin softwood and soft-hardwood species derived from mass sampling--phase I

Table 5.--Comparativespecific gravity data from other studies for 10 species sampled in the Wisconsin Wood Density Survey

(Page 1 of 2)

FPL 202

-18-

Table 5.--Comparativespecific gravity data from other studies for 10 species sampled in the Wisconsin Wood Density Survey--continued

1Samples included clear wood sticks, cross sections, and increment cores. 2(P) = plantation stock. 3Calculated from ranges. 4Values for last 10 rings (nearest cambium). 5Range of averages, not individuals. (Page 2 of 2)

-19-

LITERATURE CITED

1.

Baker, G. 1967. Estimating specific gravity of plantation-grown red pine. 107(8): 21-24.

2. Benson, M. K., and Einspahr, D. W. 1967. Early growth of diploid and triploid hybrid aspen.

Forest Prod. J.

Forest Sci. 13(2): 150-155.

3.

Born, J. D. 1966. Specific gravity of increment cores from interior Alaska. U.S. Forest Serv. Res. Note NOR-19. Institute of Northern Forestry, Juneau, Alaska.

4.

Chang, C. I., and Kennedy, R. W. 1967. Influence of specific gravity and growth rate on dry wood production in plantation-grown white spruce. Forest. Chron. 43(2): 165-173.

5.

Clark, Alexander III, and Saucier, J. R. 1969. Wood density surveys of the minor species of yellow pine in the eastern United States. Part III: Table-mountain pine. U.S. Forest Serv. Res. Pap. SE-52. Southeast Forest Exp. Sta., Asheville, N.C.

6.

, and Taras, M. A. 1969. Wood density surveys of the minor species of yellow pine in the eastern United States. Part II: Sand pine. U.S. Forest Serv. Res. Pap, SE-51. Southeast Forest Exp. Sta., Asheville, N.C.

7.

, and Taras, M. A. 1970. Wood density surveys of the minor species of yellow pine in the eastern United States. Part VII: South Florida slash pine. U.S. Forest Serv. Res. Pap. SE-66. Southeast Forest Exp. Sta., Asheville, N.C.

8.

, and Wahlgren, H. E. 1970. Wood density surveys of the minor species of yellow pine in the eastern United States. Part V: Virginia pine. U.S. Forest Serv. Res. Pap. SE-64. Southeast Forest Exp. Sta., Asheville, N.C.

9.

10

deMontmorency,W.H. 1965. The relationships of wood characteristics to mechanical pulping. Mag. Can. 66(6): T325-T348.

Pulp Pap.

Einspahr, D. W., and Benson, M. K. 1967. Geographic variation of quaking aspen in Wisconsin and upper Michigan. Genet. 16(3) : 106-112.

11. Freese, F. 1967. Elementary statistical methods for foresters.

USDA Agr. Handbk. 317.

Silvae

87 pp.

12.

Gammon, G. L. 1969. Specific gravity and wood moisture variation in white pine. U.S. Forest Serv. Res. Note NE-99. Northeast Forest Exp. Sta., Upper Darby, Pa.

13.

Gilmore, A. R. 1968. Geographic variations in specific gravity of white pine and red pine in Illinois. Forest Prod. J. 18(11): 49-51.

14. Grigal, D. F., and Sucoff, E. I. 1966. Specific gravity variation among thirty jack pine plots. 497, 498. -20-

Tappi 49(11):

15.

Hale, J. D. 1962. Minimum requirements of defining species norms for quality of variable woods, Tappi 45(7): 538-542.

16.

Heinrichs, J. F., and Lassen, L. E. 1970. Improved technique for determining the volume of irregularly shaped wood blocks. Forest Prod. J. 20(4): 24.

17.

Honer, T. G. 1970. Bole weight to volume ratio: a constant for open-grown balsam fir. monthly Res. Notes 26(4): 37.

Bi­

18.

Keith, C. T. 1969. Resin content of red pine wood and its effect on specific gravity determinations. Forest. Chron. 45(5): 338-343.

19.

Kennedy, E. I. 1965. Strength and related properties of woods grown in Canada. Can. Dep. Forest. Publ. No. 1104. Forest Prod. Res. Br., Ottawa Lab., Quebec, Canada.

20.

Maeglin, R. R. 1966. Predicting specific gravity of plantation-grown red pine. Res. Note FPL-0149. Forest Prod. Lab., Madison, Wis.

USDA Forest Serv.

21. 1967.

Effect of tree spacing on weight yields for red pine and jack pine. Forest. 65(9) : 647-650,

J.

Maeglin, R. R., and Wahlgren, H. E. 1972. Report No. 2. Western wood density survey. USDA Forest Serv. Res. Pap. FPL 183. Forest Prod. Lab., Madison, Wis. 23.

24.

Markwardt, L. J., and Wilson, T.R.C. 1935. Strength and related properties of woods grown in the United States. Tech. Bull. No. 479. Forest Prod. Lab., Madison, Wis. Mitchell, H. L. 1958. Wood quality evaluation from increment cores.

USDA

Tappi 41(4): 150-156.

25.

Okkonen, E. A., Wahlgren, H. E., and Maeglin, R. R. 1972. Relationships of specific gravity to tree height in commercially important species. Forest Prod. J. 22(7): 37-41.

26.

Pillow, M. Y. 1952. Some characteristics of young plantation-grown red pine in relation to properties of the wood. J. Forest Prod, Res. Soc., April 1952.

27.

Pronin, D. 1971. Estimating tree specific gravity of major pulpwood species in Wisconsin. USDA Forest Serv. Res. Pap. FPL 161. Forest Prod. Lab., Madison, Wis.

28.

Risi, J., and Zeller, E. 1961. Specific gravity of the wood of black spruce (Picea mariana Mill. B.S.P.) grown on a Hylocomium-Cornus soil type. Contr. Fonds. Rech. Forest. Univ. Laval. No. 6 (Orig. not seen). Forest. Abstr. 22(5062): 642.

FPL202

-21-

29.

30.

Saucier, J. R., and Clark, A. III 1970. Wood density surveys of the minor species of yellow pine in the eastern United States. Part IV: Pitch pine. U.S. Forest Serv. Res. Pap. SE-63. Southeast Forest Exp. Sta., Asheville, N.C. , and Taras, M. A. 1969. Regional variation in specific gravity of seven pines in the southern United States. U.S. Forest Serv. Res. Pap. SE-45. Southeast Forest Exp. Sta., Asheville, N.C.

31.

W. Stone R. N., and Thorne, 196 . Wisconsin's forest resources. U.S. Forest Serv. Lake States Forest Exp. Sta. Pap. No. 90, St. Paul, Minn.

32.

Taras M. A., and Saucier, J. R. 1968. Wood density surveys of the minor species of yellow pine in the eastern United States. Part I: Spruce pine. U.S. Forest Serv. Res. Pap. SE-34. Southeast Forest Exp. Sta., Asheville, N.C.

33.

, and Saucier, J. R. 1970. Wood density surveys of the minor species of yellow pine in the eastern United States. Part VI: Pond pine. U.S. Forest Serv. Res. Pap. SE-65. Southeast Forest Exp. Sta., Asheville, N.C.

34. U.S. Department of Agriculture 1941. Climate and Man. USDA, Yearbook of Agriculture, 1248 pp., 35. U.S. Forest Products Laboratory. 1955. Wood Handbook. USDA Agr. Handbk. 72, 528 pp., 36.

U.S. Forest Service. 1965. Southern wood density survey. 1965 Status report. Pap. FPL 26. Forest Prod. Lab., Madison,

U.S. Forest Serv. Res.

37. 1965. Report No. 1. Western wood density survey. FPL 27. Forest Prod. Lab., Madison,

U.S. Forest Serv. Res. Pap.

38.

Valentine, F. A. 1961. Natural variation in specific gravity in Populus tremuloides in northern New York. Proc. 9th NE Forest. Tree Impr. Conf., New Haven, Conn., 17-24 Aug. 18, 19, 1960,

39.

Wahlgren, H. E., Baker, G., Maeglin, R., and Hart, A. C. 1968. Survey of specific gravity of eight Maine conifers. USDA Forest Serv. Res. Pap. FPL Forest Prod. Lab., Madison,

40.

41.

42.

, and Fassnacht, D. L. 1959. Estimating tree specific gravity from a single increment core. Prod. Lab. Rep. 2146. Forest Prod. Lab., Madison, , Hart, A. C., and Maeglin, R. R. 1966. Estimating tree specific gravity of Maine conifers. Pap. FPL 61. Forest Prod. Lab., Madison, Wis.

U.S. Forest Serv. Res.

, and Schumann, D. R. 1972. Properties of major southern pines: Part 1--Wood density survey. Serv. Res. Pap. FPL 176. Forest Prod. Lab., Madison, -22-

U.S. Forest

USDA Forest

43.

44.

Wilde, S. A., and Paul, B. H. 1959. Growth, specific gravity, and chemical composition of quaking aspen on different soil types. U.S. Forest Prod. Lab. Rep. 2144. Forest Prod. Lab., Madison, Wis. , Paul B. H., and Mikola, P, 1951. Yield and quality of jack pine pulpwood produced on different types of sandy soils in Wisconsin. J. Forest. 49(12): 878-881.

45. Young, H. E., Hoar, L., and Ashley, M. 1965. Weight of wood substance for components of seven tree species. 466-469.

FPL 202

-23-

Tappi 48(8):

APPENDIX The purpose of the three phases of this study, as noted in the text, were: Phase I - mass increment core sampling

Phase II - increment core and data processing

Phase III - regression development to predict tree specific gravity from increment core

specific gravity. The following information details the methods used to achieve the purposes of the study. PHASE I Sample Plot Location A triple sampling design was used to establish forest survey plots across the State. First, 320,815 points distributed systematically across aerial photos using a dot-grid overlay system were observed. These points were classified as either forest land (139,867) or non­ forest land (180,948). Next a random selection of one-third of these forest points (46,104) were stereoclassified as to forest type. From these stereoclassified forest points a sample of 6,419 along with 7,960 nonforest points were examined on the ground to correct for errors in classification and for changes in land use since the photographs were taken. This ground checking resulted in 6,017 commercial forest land locations being established. For the northeast, northwest, and central survey units (fig. A1), each ground plot represents about 2,380 acres of commercial forest land. For the southwest and southeast survey units, each plot represents about 2,600 acres. Increment Core Collection At each commercial forest land location, 10 variable-radius plots (37.5 basal area factor) were established uniformly over the sample acre. At two-thirds of these locations a supple­ mental 37.5 basal area factor plot was established by the substitute point methoda for selecting trees for wood density sampling. Trees 4.0 inches and larger in diameter, of the included species, were selected and increment cores extracted at breast height (4.5 ft.). Additional trees that were selected and bored for site index and stand age determination also were included in the core sampling. A total of 4,258 cores were collected from these sample plots. All boring was done with diameter-calibrated increment borers (24). The following data were recorded in the field: Plots Survey unit county Plot number Township

Range

Section

a

Trees Species Diameter (at breast height to 0.1 in.)

Cores Diameter (to

0.001 in.)

U.S. Forest Service, Forest Survey Handbook, FSH-4809.11, section 42.9 Substitute Points, rev. 1967.

FPL 202

-24-

PHASE Il b Core Processing As increment cores were received at the Forest Products Laboratory from forest survey, all accompanying information was checked, and the cores were saturated in water to insure green dimensions. After saturation, cores were trimmed of bark and pith. The cores were then measured for length to the nearest 0.02 inch. Cores were ovendried and weighed to the nearest 0.001 gram and specific gravity calculated.

Core specific gravity = ovendry weight (grams) ÷ 12.8754 2 x (core diameter, inches) (core length, inches)

Using the basic increment core specific gravity data, mean core specific gravity and standard error for mean core specific gravity were calculated. Mean core specific gravity =

where: C ij ni

= core specific gravity of the j

th

tree at the i

th

location and

= number of trees sampled at the it h location.

Standard error of mean core specific gravity

where: k = number of locations where samples were taken and S = sum of core specific gravities at the i th location i Combined with data from phase III the above values permit calculation of standard error of mean tree specific gravity.

bAt the Forest Products Laboratory E. Arnold Okkonen was in charge of core processing and data

compilation; Mrs. Gerri Lorberter was in charge of computer processing. -25-

Standard

error

of mean

tree

specific

gravity

=

where: n

=

1

n2 X1i

the number of observations used to estimate the relationship between a Y-variable (tree specific gravity) and one or more X-variables, = the number of observations used to estimate the population mean values of the X-variables, = the mean value of Xi from which the equation was developed,

= the mean value of Xi used in the equation to predict mean Y, X2i S2 = the residual mean square of the prediction equation: y•x

2 Sy

= the variance of Y in

c

th = the ij element of the

ij

products

(of

the sample

the X'S)

used

in developing the prediction equation and

inverse of the matrix of corrected sums of squares and in the

first

sample,

PHASE IIIc Sampling

Procedure

The intensity of sampling was made roughly proportional to the volume of standing timber for each species and survey unit, as reported in the 1961 Wisconsin's Forest Resources (31). The sampling intensity was as shown in table A1, with locations by county as shown in figure A1. Field

Collection

At each location trees were sampled from diameter classes, i.e., all conifers (except plantation red pine) 4.6-5.9, 6.0-7.5, 7.6-8.9, 10.6+; aspens 4.6-6.5, 6.6-8.5, 8.6-10.5, 10.6-12.5, 12.6+. Plantation red pine sampling was by random selection (20) including eight trees of less than 4.0 inches in diameter. One increment core, to pith, was extracted at breast height from each tree selected. The cores were measured and lengths recorded to the nearest 0.02 inch. Core diameters were recorded as the borer diameter, measured with a taper gage, to the nearest 0.001 inch. Tree diameters were measured at breast height and recorded to the nearest 0.1 inch.

c

At FPL,

Dimitri Pronin (retired) and E. Arnold Okkonen were in charge of phase III operations.

FPL 202

-26-

The trees were felled and disks about 2 inches thick were cut at breast height and at suc­ cessive 100-inch intervals to a nominal 3-inch top. All disks were debarked, measured for diameter inside bark (d.i.b.), and sequentially numbered. The stump d.i.b. and age were measured and recorded. Total height was measured and recorded to the nearest 0.1 foot after felling the tree. Laboratory Procedure Core specific gravity was determined as noted under phase II; however, green core length was measured in the field for phase III. Disk volumes were obtained, after water-saturating the disks, using the rapid waterimmersion method (16). Disks were then ovendried, weighed to the nearest 0.01 gram, and specific gravity calculated. The reconstructed tree specific gravity was determined by averaging the values for the end disks of each bolt and weighting by the volume of each log (40). A more complete description of procedures is given by Pronin (27). Plantation

Data

During phase I collections, 71 jack pine, 25 red pine, and four white pine trees were sampled in plantations. Data from these trees were analyzed using unpaired "t" tests to determine if the core specific gravity means of the plantation data differed from those of the natural stands. The tests showed no difference between plantation and natural jack pine and white pine. But a significant difference was found between the specific gravity means for natural and plantation grown red pine. Because of these differences the development of separate re­ gressions was felt necessary for red pine. The data collected by Pronin (27) were all from natural stands and therefore sufficed for that population. To develop the equations for the plantation material, data from Maeglin (20) were used. The "all diameter" equation is that reported in Maeglin's study, and the raw data from the study were used to develop the equations for diameter classes (table 1).

-27-

Figure A1.--Forest survey units in Wisconsin and phase III sampling locations by species and counties within Forest Survey Units. Species, identified by number, are: 1, balsam fir; 2, tamarack; 3, white spruce; 4, black spruce; 5, jack pine; 6, red pine (natural); 7, red pine (plantation); 8, white pine; 9, eastern hemlock; 10, bigtooth aspen; and 11, quaking aspen.

-28-

Figure A2.--Balsam fir averages by unit and county. In each instance the top number (three digits) is average core specific gravity; the second is average d.b.h.; third is average age; and bottom (or right) is number of trees.

FPL 202

-29-

Figure A3.--Tamarack averages by unit and county. In each instance the top number (three digits) is average core specific gravity; the second is average d.b.h.; third is average age; and bottom (or right) is number of trees.

-30-

Figure A4.--White spruce averages by unit and county. In each instance the top number (three digits) is average core specific gravity; the second is average d.b.h.; third is average age; and bottom (or right) is number of trees.

-31FPL 202

Figure A5.--Black spruce averages by unit and county. In each instance the top number (three digits) is average core specific gravity; the second is average d.b.h.; third is average age; and bottom (or right) is number of trees.

-32-

Figure A6.--Jack pine averages by unit and county, In each instance the top number (three digits) is average core specific gravity; the second is average d.b.h.; third is average age; and bottom (or right) is number of trees.

FPL

-33-

Figure A7.--Red pine (natural) averages by unit and county. In each instance the top number (three digits) is average core specific gravity; the second is average d.b.h.; third is average age; and bottom (or right) is number of trees.

-34-

Figure A8.--Red pine (plantation) averages by unit and county. In each instance the top number (three digits) is average core specific gravity; the second is average d.b.h.; third is average age; and bottom (or right) is number of trees.

FPL 202

-35-

Figure A9.--White pine averages by unit and county. In each instance the top number (three digits) is average core specific gravity; the second is average d.b.h.; third is average age; and bottom (or right) is number of trees.

-36-

Figure A10.--Eastern hemlock averages by unit and county. In each instance the top number (three digits) is average core specific gravity; the second is average d.b.h.; third is average age; and bottom (or right) is number of trees.

FPL 202

-37-

Figure All.--Bigtooth aspen averages by unit and county. In each instance the top number (three digits) is average core specific gravity; the second is average d.b.h.; third is average age; and bottom (or right) is number of trees.

-38-

Figure A12.--Quaking aspen averages by unit and county. In each instance the top number (three digits) is average core specific gravity; the second is average d.b.h.; third is average age; and bottom (or right) is number of trees.

FPL 202

-39-

Table A1.--Phase III sampling intensity by species

1

Plantation

diameter. 4.0 in.

red pine data originated with Maeglin (20) and trees were not selected by The all-diameters regression used 66 samples, 8 of which were below diameter.

FPL 202

-40_ U. S.

Government Printing Office: 1973-754-325/21

3.5- 41-6- 73