-I

~r

I

I

!.

DESIGN
OF A
S

AND

DEVELOPMENT

N4ULTIPURPOSE

FOREST PROJECTION SYSTEM FOR
SOUTHERN

FORESTS

Bulletin 603
January 1990 Alabama Agricultural Experiment Station

Lowell T. Frobish, Director
Auburn University Auburn University, Alabama

CONTENTS
Page
INTRODUCTION............................................3 DATA....................................................5 METHODS................................................6 7 CLUSTER ANALYSIS ....................................... CROWN RATIO............................................8 DIAMETER INCREMENT......................................9 BOLE LENGTH............................................10 SURVIVAL/MORTALITY....................................11 TWIGS .................................................. 12 12 VOLUME EQUATIONS ...................................... CONCLUSIONS AND RECOMMENDATIONS......................13 14 ACKNOWLEDGMENTS ...................................... 15 LITERATURE CITED ....................................... APPENDIX 1. RELATIONSHIP AMONG SITE CLASS, 17 AGE, AND HEIGHT ...................................... APPENDIX 2. LIST OF COUNTIES BY GATWIGS............. 19 PHYSIOGRAPHIC REGIONS As USED APPENDIX 3. SPECIES GROUPS, CLUSTERS, AND MODELS....... 21 43 APPENDIX 4. VOLUME EQUATIONS .......................... APPENDIX 5. SPECIES CODE, COMMERCIAL TREES............. 47 50 APPENDIX 6. FOREST TYPE DEFINITIONS .....................

IN

FIRST PRINTING 3M, JANUARY 1990

Information contained herein is available to all without regard to race, color, sex, or national origin.

Design and Development of a
Multipurpose Forest Projection System

for Southern Forests

1

ROGER K. BOLTON and RALPH S. MELDAHL 2

SINTRODUCTION IMILARITIES in physiographic regions and forestry land use exist among states in the South. Therefore, data about forest growth processes from one state may be applicable to similar areas in other states of the region. Data from Georgia were used in development of the forest growth projection system reported herein, but the system should be applicable to similar physiographic areas in Alabama and other Southern States. Georgia has a total land area of 37.3 million acres, of which 64 percent is classified as commercial forestland (11). Within Georgia, five physiographic regions exist, as shown by the accompanying map: Lower Coastal Plain, Upper Coastal Plain, Piedmont, Ridge and Valley, and Blue Ridge. These regions are occupied by six major forest types and a variety of tree species. Due to the size and diversity of the forest resource found within the State, the development of growth and mortality models for Georgia was formidable. When developing models, this size and diversity had to be considered. Thus, a large dataset was required to model the numerous conditions represented. This report documents the modeling process used to derive a forest growth projection system for Georgia. This system is an individual tree distance independent simulation package for predicting tree and stand growth (7). Models were developed to predict live crown ratio, annual diameter increment, bole length, and mortality. These models
'Project jointly funded by Alabama Agricultural Experiment Station, Forest Resources System Institute (FORS), and Georgia Forestry Commission. 'Research Associate and Assistant Professor of Forestry.

ALABAMA AGRICULTURAL EXPERIMENT STATION A Al A~AIIA A~_e)l~

Lower Coastal Plain Piedmont Blue Ridge Upper Coastal Plain

Physiographic regions of Georgia.

were then implemented into the TWIGS 2.0 framework (3). Only the major results of these efforts are presented. If more information is desired, refer to Bolton and Meldahl (2). The development of growth and mortality models is an ongoing process. With each new model, many different statistical techniques were investigated to insure that the growth processes were adequately modeled. In addition, several new techniques were developed and utilized in the modeling process.

MULTIPURPOSE FOREST PROJECTION SYSTEM

DATA

The data for this study were obtained from the U.S. Forest Service Renewable Resources Evaluation Project. The data were from the Forest Survey of Georgia and were collected by the Southeastern Forest Experiment Station over a 3-year period beginning in 1980. In this survey, plots were systematically located on a 3-mile grid with a random start in each county of the State. These plots were based on a 10-point cluster design, with a distance of 70 feet between points. On commercial forest land, timber volume recordings were made on only three of the points. On these points, a fixed and a variable radius plot were established. The variable radius plots were used for all trees greater than or equal to 5 inches dbh, and were based on a basal area factor of 37.5. The fixed area plots were used for all trees smaller than 5 inches dbh and were based on a radius of 6.8 feet. For more detailed information on the Survey, please refer to Tansey (11). Forest Survey collects a multitude of information. As with most large sets of data, considerable time was spent examining the data and understanding variable definitions. For the purposes of modeling growth, some survey data variables are defined in ways which prohibit their use or make them difficult to use in analyses of growth. In addition, two other problems are inherent in using Survey data: Stand history is never completely known, and the sample design used was not optimal for modeling growth. However, these weaknesses in the data were not severe enough to preclude their use, and survey data provide a reasonable representation of current forest lands in Georgia. Furthermore, no other data have been found which cover the numerous growing conditions found throughout the State. Competition variables are an integral part of a distanceindependent projection system. They consist of such variables as basal area larger, trees per acre larger, number of trees per acre, and basal area per acre. The competition variable basal area larger is defined as the amount of basal area per acre greater than or equal to the basal area of the current tree. Similarly, trees per acre larger is defined as the number of trees per acre that are greater than or equal to the current tree. During the 10-year span that these measurements covered, many trees were cut, died, or grew on to the plot. Competition variables were thus needed that would measure
the effect of these trees leaving and entering the plot. Therefore, for the purposes of modeling, those trees which were ongrowth/ingrowth 1/2 of their time 1 value was contributed to the time 2 com-

6

ALABAMA AGRICULTURAL EXPERIMENT STATION

petition variables. Likewise, those trees which died or were cut were allowed to contribute 1/2 of their time 1 values to their time 2 competition variables. For example, if a tree died, then half of what that stem represents in basal area/acre at the initial measurement (time 1) was figured into the competition variables (i.e. basal area larger) for the second remeasurement period (time 2). One of the more important variables used in growth modeling is some measure of site quality. Unfortunately, site index is not directly collected by Forest Survey. Instead a site is placed into site classes, based on the potential yield of that site. This is calculated by measuring the total height and age of a dominant tree which represents the sample location. Site class is then determined by reading the appropriate graph, Appendix 1. This site class can be converted to a site index figure using the table in Appendix 1. This was not a very accurate way to measure site index, but necessary for using these data. The paucity of information dealing with natural stands in the Southeast leaves few alternatives to this approach. Physiographic regions were based on maps presented by Walker and Perkins (15), and which are currently being used by the Georgia Forestry Commission. Because the most detailed location variable within these data was county, physiographic regions were broken down according to counties. A listing of counties assigned to regions for the purposes of modeling is presented in Appendix 2.

METHODS

Four different sets of models for individual trees were developed to predict tree growth. Models were developed to predict crown ratio, annual diameter increment, bole length, and mortality. Four assumptions on tree growth were adhered to in the development of the methodology for deriving these models. First, species grow differently on the same site, e.g., loblolly pine versus sweetgum on the same site. Second, individual species may grow differently in the different physiographic regions of Georgia, e.g., loblolly pine in the Ridge and Valley versus loblolly pine in the Lower Coastal Plain, given identical stand conditions (age, site, density, etc.). Next, an individual tree may grow differently in different forest types, e.g., loblolly pine in the loblolly pine forest type versus loblolly pine in the oak-pine forest type, given identical stand conditions (age, site, density, etc.). Finally, species could be separated into three major divisions: pines, oaks, and non-oaks. Any of these assump-

MULTIPURPOSE FOREST PROJECTION SYSTEM

7

tions may be argued; however, it was felt that they provide a reasonable representation of tree and stand growth. Imposing the above assumptions on the data resulted in a rather large four dimensional matrix [physiographic region (5 regions), forest type (18 types), species division (3 divisions), and species ( > 75 species)]. It was not desirable or practical to develop models for all cells within this matrix. Therefore, a clustering analysis was developed to shrink this matrix. This clustering algorithm is a heuristic attempt to group like cells of the large matrix together. Next, regression analysis was used to derive a mathematical model for the characteristic being modeled (crown, diameter increment, bole length, mortality). This process first involved running a stepwise regression procedure to select variables for possible inclusion in the model for the aspect of growth being modeled. Then through repeated efforts to reduce the residual sum of squares and from examination of plots of residuals, a model form was selected for each major species division of the data (pines, oaks, non-oaks). Because cluster analysis also separates unlike groups, model forms can vary for clusters within each separation.

CLUSTER ANALYSIS The goal of cluster analysis is to combine similar groups and separate those which are dissimilar. Thus, for the characteristic of interest, the elements within a cluster should be more similar to each other than they are to elements of another cluster. Use of cluster analysis for grouping species was described by Meldahl et al. (8). For more discussion on the principles of cluster analysis, please refer to Everitt (6) and/or Turner (12). This clustering process was conducted for each model and species division. It starts by identifying the smallest feasible cell of the matrix, or species groups (SPG). A SPG contained at least 30 observations and/or exhibited a good distribution of the variable being modeled. These groupings were considered the smallest possible units in the clustering process. The reductions were rather arbitrary, but were based on keeping a species within similar forest types (see Appendix 6 for forest type definitions) within physiographic regions. The first step in this reduction was to find species within the five different physiographic regions which had enough observations

ALABAMA AGRICULTURAL EXPERIMENT STATION

within a forest type to be considered a separate SP G. Next, within a physiographic region, the pine forest types (FTYPE < 40), the upland oak forest types (40 < = F_TYPE = > 50), or the hardwood forest types (F TYPE > 50) were grouped together. When these groupings were not possible, either all the non-pine forest types (FTYPE > 40) were grouped together, all forest types were grouped together, or physiographic regions were combined. These species groupings attempted to combine forest types which exhibited similar growth patterns. The procedure used to group SPG's into clusters was a fourpart process: 1. Stepwise regression, plotting of variables, a literature search, and "trial and error" were used to select variables which exhibited a strong linear relationship with the function (crown class, diameter increment, bole length, or mortality) being modeled. For each of the SP G's, this simple linear model is fitted. The regression coefficients (Bo, B1) for each of these models are then standardized and entered into the clustering algorithm. 2. Cluster analysis of SPG's into 15, 10, and 5 clusters was then performed using the clustering algorithm from Proc FASTCLUS (10). The simple linear relationship which produces the coefficients which minimized the mean square error was then chosen as the "clustering function." 3. Using a graph of the cubic clustering criterion, a graph of the clusters, and through some "trial and error," an "optimal" number of clusters was obtained. 4. Final adjustments were then made to insure SP G's were intuitively correct, e.g., cedar is not assigned to the same cluster as loblolly pine. CROWN RATIO

Crown ratio is strongly correlated with stand density, competition, and tree vigor. Therefore, it is an important variable in predicting tree growth. Southeastern Forest Survey defines crown ratio as "the percent of total tree height that supports green, live foliage that is effectively contributing to tree growth" (12). This was measured for all live trees using 10 crown ratio classes, with each class representing different percents of live crown. These crown ratio classes were treated as "pseudo-continuous" variables. This was done by assigning the crown ratio (CROWNR) to be the midpoint of the crown ratio class. Since CROWNR only existed for

MULTIPURPOSE FOREST PROJECTION SYSTEM

trees at the second measurement, two sets of models were developed. One set predicted CROWNR in the future as a function of initial conditions and the other predicted CROWNR as a function of the current conditions. The first set of models was used in the prediction of growth and mortality. The latter was necessary for the prediction of bole length. Clustering of species groups was done for the three major groups (pines, oaks, non-oaks) by the methodology presented previously. Crown ratio was viewed as being very species specific. Therefore, the primary goal in deriving SPG's was to group a species together and avoid clumping of different species as much as possible. Final clusters were based on the relationship of CROWNR = f(annual diameter increment). The "optimal" number of clusters was found to be 8 for pines, 13 for oaks, and 21 for the non-oaks. Multiple linear regression analysis was then used to predict CROWNR. Clusters and models may be found in Appendix 3.

DIAMETER INCREMENT In most tree simulation packages, basal area increment or diameter increment is typically used as the basic measure of growth. West (16) showed that little difference existed between the two types of models. Since diameter growth was the main focus of interest, more flexibility and versatility were foreseen in developing a diameter increment model. Efforts were therefore directed toward developing annual diameter increment (DINC) models. The clustering technique used for diameter increment followed that mentioned previously. However, in assigning SPG's, more attention was given to separating species by region. The following variables along with the "optimal" number of clusters were used to model DINC. Group Function Pine DINC = In(predicted CROWN Oak DINC = sqrt(trees/acre larger) Non-oak DINC = ln(dbh) R) Optimal no. 12 23 18

These final clusters may be found in Appendix 3. Initial attempts at modeling DINC involved the use of non-linear regression models similar to those developed in the Lake States for STEMS (13). Such models are intellectually appealing. However, this approach did not produce satisfying results. Therefore, models

10

ALABAMA

AGRICULTURAL

EXPERIMENT

STATION

for each cluster were constructed by simply using multiple linear regression. Next, an iterative method of fitting a power transformation to DINC was used to improve these initial models. This approach to fitting a transformation follows that suggested by Draper and Smith (5). The power transformation used was: If lambda = 0 then W = ln(DINC) else W = ((DINC**lambda) - 1.0)/lambda This transformation assumes that DINC > 0. Within each group (pine, oak, non-oak), several clusters were identified for which transformations on the previously derived models had a marked impact on the analysis of the residuals. After comparing the maximum lambda value and its corresponding confidence interval for these select clusters, an acceptable value was chosen for lambda. The regressions were then estimated again using this value for lambda to calculate the power transformation. In all but a few cases, gains of 1-10 percent were noticed in the r-square. Furthermore, the analysis of the residuals showed marked improvement in every regression. The models, coefficients, and transformations for the final models are given in Appendix 3.

BOLE LENGTH Equations were needed to predict some form of tree height, for use in calculating volumes. The preferred approach would have been to develop a height increment or bole increment equation, similar to that of DINC. However, the only measure of height in the data was bole length (BOLE) of trees at time 2 which had a dbh > = 5 inches. The Southeastern Forest Survey defined BOLE as "the distance on the main stem from a 1-foot stump to a 4-inch diameter outside bark." This definition set restrictions on the volume equations used in TWIGS. The procedure used to develop the bole length models was the same as those used for DINC and CROWN__R. The cluster analysis resulted in clustering on the following equations and number of clusters: Group Function Pine BOLE = (site index*dbh)/100.0 Oak BOLE = site index*dbh Non-oak BOLE = dbh Optimal no. 14 13 20

MULTIPURPOSE

FOREST

PROJECTION

SYSTEM

11

Regression analysis was used to derive equations to predict BOLE as a function of current growing conditions. The resulting clusters, equations, and coefficients may be found in Appendix 3.

SURVIVAL/MORTALITY Mortality is one of the most difficult tree characteristics to predict. The only aspect of mortality which these data could be used to model was natural mortality. Therefore, plots which exhibited heavy mortality due to infestations of insects, disease, or fire were identified and deleted. Additionally, the nature of survey data is such that it is not known when in the remeasurement period a tree died. This inadequacy in the data severely limited the scope of the analysis. The clustering technique used for mortality differed from that used in the previous models. This technique was based on the principle that greater mortality is anticipated for trees which grow slowly than for trees which are growing vigorously. Therefore, vigor was defined by the predicted annual diameter increment, or PDINC. In deriving SP_G's upon which to perform cluster analysis, efforts were then made to group species in a region which had a distribution of dead trees across the range of PDINC. This follows recommendations by Buchman et al. (4). In order to group species further, they were placed in the following categories: RO (red oaks) - SP = 806, 812, 813, 817, 820, 827, 828, 830, 831, 833, 834, 837,838. WO (white oaks) - SP = 802, 822, 823, 832, 835 RS (scrub oaks) - SP = 807, 819, 824, 825, 840, 841, 899. SOFT (soft hardwoods) - SP = 221, 222, 313, 316, 460 555, 580, 611, 621, 651, 652, 653, 691, 693, 694, 721, 731, 740, 762, 920, 950, 970. HARD (hard hardwoods) - SP = 311, 318, 370, 400, 491, 531, 540, 552, 591, 602, 680, 901. Also, due to the enormous number of small trees which existed in the mortality dataset, only trees with a dbh > = 5.0 inches were used in deriving SP G's and in the dcluster analysis. Cluster analysis was then performed on the results of non-linear regression, by max-

12

ALABAMA AGRICULTURAL EXPERIMENT STATION

imum likelihood method, to estimate a logistic function weighted by the initial dbh. Each group (pine, oak, non-oak) used coefficients for the function mortality = f(trees per acre larger) on which to cluster. The final number of clusters was selected by examining the graphical results of the cluster analysis and by minimizing the mean square error. Logistic regression was then used to derive equations which predict the probability of survival. Initial attempts at modeling mortality utilized the full range of the data. However, the best results came from dividing the data into trees with dbh's > = 5.0 inches and trees with dbh's < 5.0 inches. The clusters, equations, and coefficients may be found in Appendix 3. TWIGS The growth and mortality models have been implemented into the TWIGS 2.0 framework. TWIGS was originally developed by the U.S. Forest Service North Central Forest Experiment Station (1). It is a menu driven program which allows the user to explore several silvicultural and economic alternatives. The program is designed for IBM PC's and compatible machines. Little was changed in this program during model implementation. For more information on running this version of GATWIGS, refer to Bolton and Meldahl (3), and for a more detailed description of TWIGS see Miner et al. (9). VOLUME EQUATIONS Volume equations which represent the forest of Georgia were an essential requirement of the projection system. However, a scarcity of volume information, especially for southern pines, and the limitation of predicting bole length severely restricted the selection and implementation of volume equations into TWIGS. In GATWIGS, volumes are calculated for the stem and product for all trees with a dbh > = 5 inches. The stem volume is measured by green weight of the bole, including bark, to a 4-inch top (d.o.b.). Volumes for sawtimber and pulpwood are also estimated. Sawtimber is defined as any stem with a bole length > = 16 feet, and for pines a dbh = 9 inches, or for hardwoods a dbh > = 11 inches. Volume is estimated for sawtimber in cubic feet and board feet. Board foot volumes are calculated by the Scribner log rule for pines and by the Doyle log rule for hardwoods. Pulpwood is defined as any stem

MULTIPURPOSE FOREST PROJECTION SYSTEM

13

that does not meet the requirements for sawtimber. It is measured in cubic feet and in cords. Cubic foot volumes for both products are only predicted for solid wood. Attempts were made to use the most recent and complete volume information for each species. Selection of a volume equation proceeded with the following steps until available equations were found in the literature. 1. If an equation existed for that species in that physiographic region which calculated volume = f(dbh, bole). 2. If an equation existed for that species in an adjacent physiographic region which calculated volume = f(dbh, bole). 3. If an equation existed for that species in Georgia which calculated volume = f(dbh, bole). 4. If an equation existed for that species which calculated volume = f(dbh, total height). Where total height is predicted using unpublished Forest Survey equations. 3 5. A miscellaneous category for that particular physiographic region. The equations finally selected and implemented into TWIGS are detailed further in Appendix 4.

CONCLUSIONS AND RECOMMENDATIONS Georgia contains a large land base and a diverse forest resource. This diversity made the modeling of forest growth a complicated task. To capture and control this diversity, cluster analysis was used frequently in the modeling process. Four sets of models were calibrated to predict growth: crown ratio, diameter increment, bole length, and mortality. In deriving these equations, statistical techniques such as multiple linear regression, non-linear regression, power transformations, and logistic regression were used. These models will generally predict growth and mortality on the "average" very well. However, the forest ecosystem is a very complex system that is difficult to mathematically model. Therefore, these models have limits and may at times appear to behave illogically. At this time, these limits have not been fully investigated or defined. Users should be careful with the inputs to the model, and use their forestry background to assess the final results. Properly used, these models should allow many different aspects of forest growth and management to be investigated and studied.
Personal communication, December 1987, J.P. McClure, FIA, Southeastern Forest Experiment Station, Asheville, North Carolina 28804.

14

ALABAMA AGRICULTURAL EXPERIMENT STATION

ACKNOWLEDGMENTS Many people assisted in the development of this projection system. However, several individuals have made major contributions. Marian Eriksson developed much of the theory and methodology of the clustering analysis; and Chuck Warlick developed many of the programs used in data manipulation and variable selection. Mike Watson and Joseph Yu spent many hours making the needed computer programming modifications to TWIGS. Appreciation is also expressed to the Georgia Forestry Commission and Forest Resource Systems Institute for their funding and support of this project.

MULTIPURPOSE

FOREST

PROJECTION

SYSTEM

15

LITERATURE CITED
(1) BELCHER, D.M. 1982. TWIGS: The Woodsman's Ideal Growth Projection
System. In: Microcomputers, a New Tool for Foresters. Proc. of a conf. sponsored by Purdue Univ. Dept. of Forestry and Natural Resources and the S.A.F. Systems Analysis and Inventory Working Groups. (Purdue University, West Lafayette, Ind., May 18-20, 1982.) (2) BOLTON, R.K AND R.S. MELDAHL. 1986. A Multipurpose Forest Projection System for Southern Forests. Final Report for Georgia Forestry Commission Cooperative Agreement. . 1990. User's Guide to a Multipurpose (3) Forest Projection System for Southern Forests. Ala. Agr. Exp. Sta. Bull. 604. (4) BUCHMAN, R.G., S.P. PEDERSON, AND N.R. WALTERS. 1983. A Tree Survival Model with Applications to Species of the Great Lakes Region. Can. J. For. Res. 13:601-608. (5) DRAPER, N.R. AND H. SMITH. 1981. Applied Regression Analysis. 2nd edition. John Wiley and Sons, New York, N.Y. pp. 225-241. (6) EVERITT, B. 1980. Cluster Analysis. 2nd edition. Halstead Press, New York, N.Y. (7) MELDAHL, R.S. 1986. Alternative Modeling Methodologies for Growth and Yield Projection Systems. In: Proceedings of a Computer Conference and Third Meeting of the Forest Resource Systems Institute, April 7-9, 1986, Atlanta, Ga. Forest Resource System Institute. Florence, Ala. pp. 35-39. (8) , M. ERIKSSON, AND C.E. THOMAS. 1984. A Method for Grouping Species-Forest Type Combinations for the Development of Growth Models for Mixed Species Stands. In: Proceedings, Third Biennial Southern Silvicultural Research Conference. (E. Shoulders, Ed.) USDA Gen. Tech. Rep. SO-54. p. 422-442.
(9)
MINER, C.L., N.R. WALTERS, AND M.L. BELLI. 1988. A Guide to the

TWIGS Program for the North Central United States. USDA For. Ser. N. Cen. For. Exp. Sta., St. Paul, Minn., Gen Tech Rep. NC-125. (10) SAS INSTITUTE INC. 1982. SAS User's Guide: Statistics. SAS Institute Inc., Cary, N.C. pp. 433-447. (11) TANSEY, J.B. 1983. Forest Statistics for Georgia, 1982. USDA For. Ser. S.E. For. Exp. Sta. Res. Bull. SE-69. (12) TURNER, B.J. 1974. Applications of Cluster Analysis in Natural Resources Research. For. Sci. 20:343-349.

(13)

U. S. DEPARTMENT OF AGRICULTURE. 1979. A Generalized Forest Growth

Projection System Applied to the Lake States Region. USDA For. Ser. Gen. Tech. Rep. NC-49. (14) . 1980. Field Instructions for Georgia. Renewable Resources Evaluation Project S.E. For. Exp. Sta. (15) WALKER, L.C. AND H.F. PERKINS. 1958. Forest Soils and Silviculture in Georgia. Ga. For. Res. Rep. No. 4. (16) WEST, P.W. 1980. Use of Diameter Increment and Basal Area Increment in Tree Growth Studies. Can. J. For. Res. 10:71-77.

MULTIPURPOSE FOREST PROJECTION SYSTEM

17

APPENDIX 1. RELATIONSHIP AMONG SITE CLASS, AGE, AND HEIGHT
TABLE 1. RELATIONSHIP BETWEEN SITE CLASS AND SITE INDEX' 2

Species 1 White pine. ........................... 74+ Longleaf pine ...................... 123+ Slash pine...........................103 + Loblolly pine ....................... 110+
Shortleaf pine....................... 104+

Site index range , by site class 2 3 4 58- 73 44- 57 35-43 106 - 122 89 - 105 68 - 88 93 - 102 80 - 92 62 -79 95 - 109 80 - 94 60 - 79
86 - 103 72 - 85 54 - 71

5 3467 61 59 53 -

Virginia pine....................... 79+ Pond pine....................... 133+ Sweetgum..............................--Yellow-poplar ...................... 135+
Mixed oaks.........................128+

74- 78 118 - 132 119+ 108 - 134
109 - 127

67- 73 93 - 117 90 - 118 87 - 107
89 - 108

57- 66 66 - 92 75 - 89 68 - 86
60 - 88

5665 75 67 59 -

'Personal communication, July 1986, J.P. McClure, FIA, Southeast Forest Experiment Station, Asheville, North Carolina, 28804. 2 Height in feet at age 50.

18

18 ALABAMA AGRICULTURAL EXPERIMENT STATION

Red Gum

White Cedar-

Yellow-poplar 10-

Mixed Oaks

120
2

100 80 60 40 20
30 50 70 30
50 70 30 50 70 30 50 70

5

4 5

5 5

White Pine

Pond Pine
_

[ongleaftPine
-

120
C

1

_

-1

j o-

OF

E
0

100 80 60 40 20
Loblolly Pine-

/

,

3

'5
0)

-' 50

4

0)
O-

70

30 50 30

..

70

30 50 70
--

Shortleaf Pin~e

-

Vi inia Pine-

Slash Pine
1

120
2z

100 80 60 40 20
30

2

'12o
3O_
_0_-13 2 4 5
3

01

4

~55
513 /1303U 5130/0

513z13

11U0 13,/00eU

Age

Age

Age

Age

Site class on culmination of mean annual growth
Site class based on culmination of mean annual growth (14).

MULTIPURPOSE

FOREST

PROJECTION

SYSTEM

19

APPENDIX 2. LIST OF COUNTIES BY PHYSIOGRAPHIC REGIONS AS USED IN GATWIGS
Blue Ridge Fannin Gilmer Rabun Towns Union

Lower Coastal Plain
Appling Atkinson Bacon Ben Hill Berrien Brantley Brooks Bryan Bulloch Burke Camden Candler Charlton Chatman Clinch Coffee Colquitt Cook Dodge Echols Effingham Emanuel Evans Glynn Grady Irwin Jeff Davis Jenkins Johnson Lanier Laurens Liberty Long Lowndes McIntosh Montgomery Pierce Screven Tattnall Telfair Thomas Tift Toombes Treutlen Turner Ware Wayne Wheeler Wilcox Worth

Piedmont
Baldwin Banks Barrow Butts Carroll Cherokee Clarke Clayton Cobb Columbia Coweta Dawson Dekalb Douglas Elbert Fayette Forsyth Franklin Fulton Greene Gwinnett Habersham Hall Hancock Haralson Harris Hart Heard Henry Jackson Jasper Jones Lamar Lincoln Lumpkin Madison McDuffie Meriweather Monroe Morgan Newton Oconee Olgethorpe Paulding Pickens Pike Putnam Rockdale Spalding Stephens Talbot Taliaferro Troup Upson Walton Warren White Wilkes

20

ALABAMA AGRICULTURAL EXPERIMENT STATION

Upper Coastal Plain
Baker Bibb Bleckley Calhoun Chattahoochee Clay Crawford Crisp Decatur Dooly Dougherty Early Glascock Bartow Catoosa Chatooga Dade Houston Jefferson Lee Macon Marion Miller Mitchell Muscogee Peach Pulaski Quitman Randolph Richmond Schley Seminole Stewart Sumter Taylor Terrell Twiggs Washington Webster Wilkerson

Ridge and Valley
Floyd Gordon Murray Polk Walker Whitfield

MULTIPURPOSE

FOREST

PROJECTION

SYSTEM

21

APPENDIX 3. SPECIES GROUPS, CLUSTERS, AND MODELS Species groups (SP_G) were assigned names based on a three part naming convention. The first part of the name identifies the physiographic region, the second part the species, and the third part the forest type. A typical SPG name is then L611F60. The first character(s) of a SP_G name represents one of the following regions: L U P V B A = = = = = = Lower Coastal Plain Upper Coastal Plain Piedmont Ridge and Valley Blue Ridge all regions

All but the last code may be combined to represent several regions. The next 3 digits are simply the Forest Service's species codes, Appendix 5. The rest of the SPG name represents the forest type by one of the following codes: FXX = forest type XX, where XX is forest type (Appendix 6) A = all forest types OP = oak - pine forest types (FTYPE < = 50) H = hardwood forest types (FTYPE > 50) O = oak forest types (FTYPE = 40 or 50) PH = mixed forest types M = miscellaneous, all forest types excluding those assigned to other groups All species not assigned to a SP_G are placed into a category called OTHER. Therefore, the SP_G name L611F60 represents sp = 611 (sweetgum) found in the Lower Coastal Plain, and located in Forest Type = 60 (oak-gum-cypress).

22

ALABAMA AGRICULTURAL EXPERIMENT STATION

The following is a listing of the variables and their computer symbols, which are used throughout the appendices. BAL T1 BAL T2 DBHT1 DBHT2 PLCR SITE STANDA STAND_AT2 TALT1 TAL T2 TOBA1 TO BA2 TOTA1 TO TA2 basal area larger - time 1 basal area larger - time 2 dbh - time 1 dbh - time 2 predicted live crown site index stand age - time 1 stand age - time 2 trees/acre larger - time 1 trees/acre larger - time 2 total basal area - time 1 total basal area - time 2 total number of trees/acre - time 1 total number of trees/acre - time 2

MULTIPURPOSE

MULTIPURPOSE FOREST PROJECTION SYSTEM
FOREST PROJECTION SYSTEM.

23

23

Table 1. Crown Ratio Clusters

Pine' Cluster 1. 2. BliQA A121A B132A P11OA L131A P131A P131P1 L131P1 U11OA
L11OA

Oak P832A V832A 8832A BVSREDA 3 PU8O6A V833A V837A V806A

Non-oak AMAGA' 2 VLMAPELA B400A AHACKA4 P694A A531A U691A V611A

3.

U131A UlilA V131A U111PL V11OA

L835A 2 PVSCRUBA A831A

4.

8833A PU837A B806AV802A
P833A B802A L802A L827A P824A LSREDA 3 U807A U835A L8fl7A PU819A L820A 3 PSREOA PVB827A U838A V835A B837A AOTHERA PUL825A P802A U802A U827A
P835A

L222A A692 U221A B621A L221A L653A L694A B316A L721A P316A B491A B7I1A PUL71 lA B693A L316A AOTHERA AO6OA P391A P400A P611A

V7L1A

5.

P540A

V612A A9O2A V400A U970A U521A V316A PVB762A U653A U694A

6.

LillA L111PL '131P1 A128A A129A P132A U131P1 V132A

7.

8.

9. 10.

PVBMAPLE2 U400A

'SP
2

SP SP SP

-

651, 652. 310-318, not including 316. 812, 813, 834. 460, 461, 462.

24

24

ALABAMA AGRICULTURAL EXPERIMENT STATION

Table 1 (continued). Crown Ratio Clusters

11. 12.

L819A PUL822A L838A P820A USCRUBA4 4 LSC RUBA U820A USREDA 3 VULB824A

A519A L512A P512A V491A V693A L391A L400A U491A L762A L693A P693A PV97OA A7O1A ABIRCHA L970A L611A L621A A680A L540A L691A A555A A931A
5

13. 14.

U316A U611A U693A P491A U391A L491A P653A. P621A U762A U540A U621A U222A VB521A

15. 16. 17.

18.
19.

20.
21.

MULTIPURPOSE

MULTIPURPOSE FOREST PROJECTION SYSTEM
FOREST PROJECTION SYSTEM

25

25

Table 2

.

Coefficients for Crown Ratio Equations

-

Pines

Cluster CROWN..R
-

BO BO
+

B1

B2

83

B4

B1*([SITE*STANDt.AJ/100.00) + B2*(USITE*TO.BA1J/100.00) + B3*(SITE*DBHT1) + B4*(ln(TAL.T1)) -0.00154260 -0.00166059 -0. 00158220 -0.00136857 0.00076801 -0.00187729 -0.00080745 -0.00277728 -0. 00085430 -0. 00101115 -0. 00035415 -0.00049365 0.00348249 -0.00006025 -0. 00095414 -0.00025979 0. 00002096 0. 00013397 0. 00000780 0. 00000614 0. 00042025 -0. 00004120 -0.00002616 0.00000336 -0.01003733 -0.00476772 -0. 03170961 -0.02745300 -0.01397368 -0. 03936502 -0. 02442492 -0.02539228

0.5867 1978 0.43379111 0.59532667 0.57858310 0.16596777 0. 61825820 0. 68482619 0.57962489

Table 3. Coefficients for Crown Ratio Equations- Oaks

Cluster CROWN_R 2 5 6 8 9 10 12 CROWN.R 3 4 99*
-

BO

B1

B2

B3
+

B4

B0 + B1*(lnLTAL.T1.]) + 82(ln[STAND .A])

B3*(ln([O..BA1])

0.61787371 0.63921987 0.75320210 0.90442649 0.66325143 1.22163682 0.76361118 0.73126488
-

-0.01616234 -0.01587070 -0.01145295 -0.01194511 -0.01179819 -0.01510610 -0.01896173 -0.00515414
+

-0.02508335 -0.03338825 -0.05660677 -0.05587069 -0.03593225 -0.20061924 -0.07481344 -0.02007368
+

0.00143028 0.00395124 0.00736719 -0.02159680 0.00884265 0.02520893 0.02180847 -0.02522286
+

80B1*(SITE)
+

B2*(TOTA1)

B3*(ln(STANDA))

B4(TOTA1/SITE) 0.01886271 -0.00790070 -0.00100139

0.71492827 0.79028349 0.68471313

0.00292671 -0.00169732 0.00068135

-0.00030155 0.00007943 -0.00000720

-0.09198565 -0.04394815 -0.02487705

*Clusters

-

7, 11, & 13 were combined into this cluster.

Table
Cluster BO CROWN_R 2 4 5 6 7 8 10 11 12 14 16 19
=

4. Coefficients for Crown Ratio Equations

-

Non-Oaks
B4 B5

B1

B2

B3

B0 + B1*(TOTA1/DBHT1) + B2(ln[STAND A]) + B3*(ln[TO TAll) + B4*(ln[TO BAli) 0. 00001585 -0.00001136 -0. 00001321 -0.00004175 -0.00004038 -0.00000718 -0.00003639 -0.00004749 -0.00011489 -0. 00001 988 -0.000066340.00000699 0.00000792 0. 00001797
=

20
21

0.65077629 0.61681247 0.71433569 0.80922492 0.36802397 0.73074453 0.73 121784 1. 24474979 0.30301827 0.8080 1309 0.58024327 0.65440445 0.58497571 1.03782183 CROWN R

-0.06493664 -0.00146183 -0.01624137 -0.02685701 -0.01651700 -0.01850771 -0.04010656 0.01095073 -0.02768460 -0. 17830 062 -0.04430699 -0.02788155 -0. 0097 1895 -0. 04827582

0.00035677 -0.00685707 -0.02497831 -0.01903849 0.00352634 -0*.03381165 -0.01568840 -0.03023905 0.04824427 -0.01391437 -0.00931027 -0.00657313 -0.00728645 10125023 -0.

0.01390576 -0.04407853 -0.01476367 -0.02114446 0.02700426 0.00237704 0.00741302 -0. 10812156 -0.00873382 0.00153.530 0.03935074 -0.01117683 -0.02832546 0.05461962

BO + B1*(STANDA) + B2*((SITE*STANDA)/100.00) + B3*(ln (TOBAl)) + B4*(sqrt(TALTi)) + B5*(SQRT(BALTi)) 0.00266222 -0.00269809 -0.00166099 -0.00748811 0.00153455 0.00007289 -0.06161525 -0.08382437 0.01325696 -0.00626863 -0.00172909 -0.00308951 0.03579465 0.02772736 -0.00178924

15 17 99*

0.75079723 0. 75718235 0.66378006

*Clusters

1, 3-, 9, 13,

&-18werecombined

into this Cluster. -

MULTIPURPOSE

MULTIPURPOSE FOREST PROJECTION SYSTEM
FOREST PROJECTION SYSTEM

27

27

Table 5. Coefficients for Crown Ratio Equations

-

Pines

Cluster

BO

BI
-0.00154260 -0.00166059 -0. 00158220 -0.00136857 0.00076801 -0.00187729 -0.00080745 -0.00277728

82

B3

B4

CROWN_R

-

BO + B1*(SITE*STAND..A]/1.00.O0) + B2*([SITE*TO..BA2]/100.00) +B3*(SITE*DBHT2) + B4*(ln[TAL.T2]) -0.00085430 -0. 00101115 -0.000354 15 -0. 00049365 0. 00348249. -0.00006025 -0.00095414 -0.00025979 0.00002096 00013397 0. 0. 00000780 0. 000006 14 0.00042025 -0.00004120 -0.00002616 0.00000336 -0.0 1003733 00476772 -0. 0.03 170961 -0.02745300 -0. 01397368 -0.03936502 -0.02442492 -0.02539228

0.5867-1978 0.43379111 0.59532667 0.578583 10 0.16596777 0.61825820 0.684826 19 0.57962489

Table 6. Coefficients for Crown Ratio Equations

-

Oaks

Cluster

BO
-

81

B2.

B3

CROWN_-R
2 4 5 6 8 9 10 12 99*

80

+ B1*(ln[TAL.T2]) + B2(ln[STAND.AT2]) + B3*(ln(TO.BA2])

0.67857626 0.63670670 0.626 17555 0.77268513 0.92013974 0.70257374 1.32784652 0.85902067 0.80082060 0.77501650

-0.0199988,1 -0.01611040 -0.00291208 -0.01482497 -0.01566130 -0.01414789 --0.01992882 -0.02427357 -0.00657727 0.00 170704

-0. 02430130 -0.0311295 1 -0.03639119 -0.05702860 -0.07268841 -0.03335369 -0.-18582340 -0. 06895190 -0.01523777 04043025 -0.

-0.00980732' 0.00177135. -0.00273879 0.006 13392 -0.00602839 -0.00073750 -0.00994076 -0.00054510 -0.04261648 -0.00997731

Clusters

7, -.

11, & 13 were combined into this cluster.

Table 7. Coefficients for Crown Ratio Equations

-

Non-Oaks

Cluster

BO

B1

B2

B3

B4

B5

CROWN_R = BO + B1*(TOTA2/DBHT2) + B2(ln[STANDAT2]) + B3*(ln[TO TA2]) + B4*(ln[TOBA2]) 2 4 5 6 7 8 10 11 12 14 16 19 20 21 0.62 18 0129 0.55825051 0.69794133 0.79805477 0.18518734 0.69166192 0.66045308 0.88998435 0.13047278 0.81262402 0.49861277 0.67675808 0.60469732 1.08977746 CROWN R 15 17 99* 0.45496672 0.83950888 0.55713463
=

0.00000346 -0.00004113 -0.00002532 -0.00007935 -0.00008614 -0.00003365 -0.00008336 -0.00019138 -0. 00023119 -0.00005201 -0.00014914 -O0.00002296 -0.00001528 -0.00000372
+

-0.05969193 -0.01687799 -0.01454281 -0.02719698 -0.01481022 -0.01234013 -0.04168485 0.02014112 -0.02430671 -0.05911250 -0.04452100 -0.03494872 0.00668729 -0.02741897

0.01126605 0.00384287 -0.01830821 -0.01107934 0.02665836 -0.01990346 -0.00322262 0.02409889 0.08510241 -0.00796262 0.00633516 -0.00236557 -0.00205668 -0.09396750

0.00090678 -0. 175050 03 -0.02131153 -0.02819772 0.03298846 -0.01304923 0.00756447 -0. 10489869 -0.01991285 -0. 00937431 0.03774567 -0.01444909 -0.05084870
0.01343640

ICI-

r

C)
C,

I-

BO + B1*(STANDA) + B2*([SITE*STANDAT2]/100.00) + B3*(ln[TOBA2])
B4*(sqrt[TAL12]) +

BS*(sqrt[BALT2])

m 0.02580512 0.02653958 -0.00572902 m z
I CO) -I

0.00570173 -0.00247987 -0.00163836

-0.01193936 0.00116878 -0.00004506

0.03 163636 -0.09304467 0.04333391

-0.00755879 -0.00317068 -0.00307408

Clusters

-

1, 3, 9, 13, & 18 were combined into this cluster.

z

MULTIPURPOSE

MULTiPURPOSE FOREST PROJECTION SYSTEM
FOREST PROJECTION SYSTEM

29
29

Table 8. Diameter Increment Clusters

Pine Cluster 1. P131PL U131F40 L827F60 P812F40 U812H V832F50 P131F50 U131F31 U11OF3O L121F22 P131F32 U121F21 L131F31 L111F21 V13IMPO B802A L838A P832A B812A

Oak

Non-oak

P812F32 P827F50 U827F50 V833A B806A P8O2MPO P835F50

BOTHERA L540F60A P316F70 U611F40 U970H P621F31

B316MPH POTHERF7O U54OMOH U693A V400P P621F50

2.

B132A P132A U131F32 L131A L111F22 P110F32 U11OF4O L131PL L111PL P131F31 UI21H P132F33 V132F33 L121MPH L110F32 P11OF5Q U1I1PL BliDA
PliOMP

3.

L621F60

4.

POTHERA U831A B833A

UOTHERA

L653M0P

U621067

5. 6. 7.

B400A P835F40 V8320 P316F60 L611F22 UOTHERMP U611F60 U621M0 POTHERP B316F50
L611F31

U131PL A129A L128MPH L111F4O

B837A U807A V802A

U200A P540F70 U611F31 U611F70 V621A P316F40 B621A
L611F50

8. 9.

L807A L111F6O

L819A P820A

P11OF31
V11OH

L812A

P11OF4O V132MPH

POTHERF5O P611F32 U400F50 V316A L840P V824A B711A L316MPO L721F60 POTHERF31 P491F50 U391A V491

P491MPO UOTHERF7O VOTHERP V611A L222F22 L591A POTHERMH P491F40 P970P U491F50

10.

L131F20 U111F22

PL32MP

30 v v

ALABAMA

AGRICULTURAL

EXPERIMENT

STATION

Table 8 (continued). Diameter Increment Clusters
11. L128F22 VI31PL U131F50 L820F40 U820F60 LOTHERF4O LOTHERF7O L611F40 POTHERF6O P3l6MPO P711F50 UOTHERF6O U540F60 VOTHERMH B491A LOTHERF5O L316F40 L653F60 LOTHERF6O L316F60 L970A P316F50 P611F50 UOTHERO U31GHP U611F50 V711A B711F50 L222F40 L391H L692H
L694F40

12.

L121F21 L111F31 P121A VllOP

L128F36 P131F40 U110F32 V131F31

L820F50

L694F22
L694F60 P540MPH P693P P970H U491P U694F40 U694F60 13. 14. L820F60 L820P V806A P827P P621H L221A L400A L721MPO L621M0 U691A LOTHERP L611F60 PO6OA P762P B693A L491A L694F50 P400F40 P521A P711F40 U400MPH V400H

P491F31 P693H P762H U491MH U521A U694F50 V693A

15. 16. 17.

L822A L827F40 U812P L827F50 P812F31 P827F40 B832A U835A

U802A P802F31 VOTHERA L827P P812F50 U827F40 P835P V837A

L316F22 L694F31 P711P P621F40

L316F50 L693A P611F60 U762A L222F60 L691F60 P391A P400MPO P694A U316F60 U653A

18.

19.

LOTHERA P802F40 V832F40

P802F50 0837A

MULTIPURPOSE FOREST PROJECTION SYSTEM Table 8 (continued). Diameter Increment Clusters
20. 21. 22. 23. L831A L835A U840A P833A L840H P824A V835A U827F60 U819045

31

Table 9. Coefficients for Diameter Increment Equations

-

Pines

Cluster

BO

Bi

B2
+

B3 B3*(DBHT1)
+

B4 B4*(TOBA1/SITE)-+ -0.03988400 -0.03898688 -0.06665636 -0.12640480 -0.38184447 0.03314018 -0.03647413 -0.15176792 -0.08749770 0.05211989 14162841 -0. -0.06934679

B5

W= BO + B1*([SITE*PLCR]/100.00) + B2*(STANDA) B5*(sqrt[TALTi]) 1 2 3 4 5 6 7 8 9 10 11 12 -0.75115908 -0.97690193 -1.06206566 -0.43816368 1.01195724 -1.23218964 -1.42595380 0.12138235 -1.10464488 -0.22525604 0.03288369 -0.86071304 0.00806429 0.00815043 0.00307892 0.00059790 -0.03956711 0.00563915 0.01619828 -0.00218810 -0.00651841 -0.00367436 02243265 -0. 0.00397704 -0.01132985 -0.00771632 -0.00567702 -0. 01077 181 0.00250204 -0. 00617363 0.00009209 -0.02551635 -0.00473196 -0. 01156928 0.00086314 -0.00560024

0. 00157051 -0.00608998 0.00252820 -0. 00714177 -0.03600948 0.01967963 -0.03658446 0.00407352 0.01464383 -0.00900284 -0.01263540 -0. 01168674

-0.02323115 -0.02174645 -0.01834552 -0.02635387 -0.04481584 -0.01863985 -0.01811980 -0.03106088 -0.01382910 -0.03895463 -0.02612898 -0.02732836

Where, DINC
=

([W*0.35]

+

1.0)

**

(1.0/0.35)

Table 10. Coefficients for Diameter Increment Equations

-

Oaks

Cluster.
W
=

BO
INTERCEPT
+

Bi

B2
+

B3
+

B4
+

B5

B1*(PLCR)

+

B2*(STANDA)

B3*(DBHT1)

B4*(TOBA1/SITE)

B5*(sqrt[TALTi])

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

-1.32501399 -0.69893090 -1.19931070 -0.79571436 -2.11679538 -1.67882541 -2. 30140428 -1.97812530 -2.70774893 -1. 91939813 1.20203445 -2.72672523 -1.33399377 19.57062466 -2.44590566 -1.91519010 -0.87210940 -0.96932027 -0.74987418 -14.99731693 -2. 12058687 -0.57903963 -2. 57541306 Where, DINC
=

0.62536478 -1.39469354 -0.01793073 -0. 81901111 2.05830852 0. 32125481 1.85481096 0.21271929 2.56578690 0.48722724 -3.70715547 3.30930023 0:65613847 -38.18846567 2.67112625 1.64066979 -0.58765289 -1.06529334 -0. 57021006 22.09654205 0. 61202956 -0.06720399 1. 21281405

-0.00298999 -0.00301444 -0.00594414 -0.00544189 -0.00296299 -0.00383395 -0.00027839 -0.00146257 -0. 00571508 -0.00605506 -0.00785543 -0.00207102 -0. 00142158 -0.04068608 -0.00164263 -0.00085211 -0.00472228 -0.00532527 -0.00578678 0. 03816865 -0. 00008411 -0. 00850916 0.00080182

-0.00477197 0.00127903 -0.00590567 0.01762343 0.02 138262 0.00840340 0.00562756 0.02 108047 0.02487778 -0.00849574 0.01748542 -0.00657351 -0.00394879 0.06684565 -0.00248346 -0. 1890 0082 0.01006620 0.00847158 -0.00310797 -0. 01793224 0.00202469 -0.00237341 -0.01050180

0.00368176 -0.01642658 0.08348777 0.02547589 -0.44587966 -0.07240059 0.00766738 -0.07086816 0.09385432 0.05257516 0.02500008 0.09670765 -0.07111605 0.20128000 0.02445698 -0. 17291823 -0.017386 12 0. 08144186 0.10140864 0.58597100 0. 00710054 0. 01434030 0.02289566

-0.03 132648 -0. 02274301 -0.03582079 -0. 02125517 -0.02055009 -0. 01Q29927 -0. 02627102 0.00636795 -0.02360776 -0. 01028813 -0.03934875 -0.03632951 -0.02428226 -0. 05169545. -0.03628908 -0.02346719 -0.02144100 -0.02260131 -0.03554657 -0.05525150 -0.01818277 -0.03577062 -0.01699124

([W*0.25]

+

1.0)

**

(1.0/0.25)

Table 11.

Coefficients for Diameter Increment Equations

-

Non-Oaks

Cluster W 1 2 3 4 5 6 7 8 9 10 11 12 '3 14 15 16 17 18
=

BO

B1

B2

B3

B4

B5

BO + B1*(PLCR) + B2*(STANDA) + B3*(DBHT1) + B4*(TOBA1/SITE) + B5*(.sqrt[TALTi]) 0.21118800 1.94210085 -16.42481019 -1.71999703 05852615 -5. -0.70216262 -0.06903728 -2.27238536 03120181 0. 0.57091372 -0.60352128 0.55564419 13.90509111 -0.06537910 -0.04823953 10.37694935 0.52192240 0.36726507 00071122 -0. -0.00626071 -0.01039097 -0.00478461 -0.01529672 00008131 -0. -0.00481095 -0.0063 1562 00417812 -0. -0.00250353 -0.00404528 -0.00153230 0.01443 199 -0.00559050 -0.00475800 0.02879308 -0.00309983 -0.00084662 0.04339527 01925349 -0. -0.00526574
0.01041694

-2.13947221 -1.42372883 7.76276 114 01147462 -0. 0.82591182 -1.63291141 20482426 -1. -0. 41299856 52526289 -1. -2.20095830 -1.41045273 28900036-2. -8.20024938 -1.34449583 -0.61982843 -5.45943363 -1.75875152 -2.39501254

0.04818579 0.03768434 0.012 14267 0.04564528 0.01428138 02521788 -0. 0.01640188 0.00991791 0.00986969 0.00635124 -0.02121113 -0.07875779 0.01508418 0.02627356

0.01381083 0.04625379 -0.40325556 -0. 15415411 0.01617225 -0.02364058 0.01873088 -0.10769231 0.05235534 0.07363077 0.02463621 0.00154364 -0.00153866 0.00256852 0.21382103 20439142 -0. -0.09567373 0.04581283

-0.01397865 -0.03884221 -0.03398598 -0.03744622 -0.01474917 -0.01524475 -0.02619186 -0.01291362 -0.02391424 -0.01035861 -0.01652395 -0.00861532 0.00161674 -0.01967158 -0.04967325 10490343 -0. -0.01377002 -0.01024107

Where, DINC = ([W*0.20] + 1.0)
**

(1.0/0.20)

MULTIPURPOSE FOREST PROJECTION SYSTEM Table 12. BOLE Length Clusters

35

Cluster
1. B1iQA P131F31 U131F40 B132A L111F4O U110F32 L121MPH U131F50 V11OH L131F31 P131F40 U131PL UL31MPO

Pine'
L128F36 U131F31 L111F22 POTHERA L131F40 POTHERH U802A POTHERA B812P P837A

Oak
P802MPH

Non-oak

BOTHERA POTHERF5O P611F60

L316MPH UOTHERF5O U611MPH

2.

POTHERF5O U827F5O

3.

L6L1MPH L694H V400A B621A B837A LOTHERF5O L820F60 P802F60 U82OMPH UOTHERA B806A P802F50 P820A P833A P835MPH P612H LOTHERMPH L691A POTHERF7O P316F70 P621F50 U62IMPH B316A P400F50 U400F50

L694F40 P693A

4.

UOTHERF5O L131MH P131F50 U121A B833A P832A B832A L820F50 L838A U820F50 L812A U812A B802A L827F60 P812F50 P827MPH P835F50 V802A L82OMPH U827MPH P827F50

5.
6.

U621F60 L316F60 L694F60 P316F50 UOTHERMPH P621P L222F40 P611F40

7. 8.

LL1F6O L131PL LiliMPO V131A

L121F21 L11IPL VilOP

9.

P132A V132A L121F22

U1100MPH

P491MPH L222MPH U653A

10.

36

36 ALABAMA AGRICULTURAL EXPERIMENT STATION Table 12 (continued). BOLE Length Clusters

11. 12.

U111A A129A

L822A L827F50 POTHERMPH VOTHERA P812F40 V832MPH P131PL

V837A L827MPH U835A V832F50

LOTHERF6O POTHERMPH P400MPH P711A L611F40 L54OA U611F40 L555A U694MPH L222F60 U619A B400A L611F50 L653A U540A P540A U611F5O U694F60 L694P P316MPH U316A L621A

L400A P611F31

13. 14. 15. 16. 17.

L128MPO Ll31MPO P131F32

P611MPH U400MPH V621A UOTHERF6O P694A L221A L611F60 P611F50 U970A P621F40 U611F60 L970A P970A VOTHERA U200A

18.

19.

20. Table 13.

Coefficients for Bole Length Equations

-

Pines

Cluster

BO BOLE LENGTH
-

B1 BO

82

B3
+

B4

+ B1*((SITE*OBHT2)/100.00) + B2*(SITE*PLCR) B3*(BAL.T2/SITE) +B4*(TO..TA2)

1 2 3 4 5 6 7 8 9 10 11 12 13 14

39.01972153 30.97143318 37.99074464 14.26990172 37.05009096 28.38612923 55.80657728 24.29234475 11.15604203 63.10088950 34.39236033 72.16468006 66.38969504 43.68922243

3.99180016 4.99140557 3.39231491 6.91213481 3.70222923 4.45845522 4.35174239 5.38883660 6.06818162 2.45991418 5.84085188 5.53988509 2_61791743 4.95560908

-0.79875633 -0.67914014 -0.6085746.1 -0.73296781 -0.65346276 -0.64896485 -1.13738949 -0.68540608 -0.49167233 -0.87090472 -1.24383871 -1.89621549 -0.94309257 -1.32360432

-0.95269580 0.18485098 1.21033063 1.86129142 1.34585696 3.60397469 -0.53589665 -0.63120150 3.80768963 -3.73184530 1.35982669 -1.94637422 -11.95175404 -2.21564176

-0.00067370 -0.00101922 0.00064769 0.00036666 -0.00168459 -0.00303813 -0.00239366 -0.00000337 0.00159825 -0.00347381 0.00405467 -0.00848680 -0.00374410 -0.00051928

Table 14.

Coefficients for Bole Length Equations

-

Oaks

Cluster

BO

Bi

B2

B3

B4

B5

BOLE LENGTH = BO + B1*(DBHT2) + B2*(SITE) + B3*(BALT2) + B4*(TOTA2) + B5*(PLCR) 1 2 3 4 5 6
7

52.50031126 22.90544722 127.87164288 -12.57362366 -10.00665124 51.98035344
7.14536603

3.03489054 2.83122631 2.25017585 2.39940996 1.97965551 1.87333317
2.65281021

0.02492904 0.20490537 0.23182976 0.27974278 0.38253461 0.08563130
0.23369684

-0.02899222 0.04241212 -0.01714174 -0.02918460 -0.00118635 0.00525169
0.05243512

-0.00309185 0.00073246 0.00130502 0.00618849 -0.00304441 -0.00193369
0.00075709

-80.85865588 -67.34047230 -279.35981767 -2.42709816 7.24451735 -74.32969075

-41.17305582
-47.09514635 81.91910065 -87.60492903 52.86717459 -48.44069205 -1.63304060

8 9 10 11 12 13

33.35573057 -13.03859804 58.56993240 -23.42788398 17.36207348 -2.06073224

2.48614798 2.29443313 2.46374607 1.54099988 3.06505208 2.54503349

0.06312505 -0.07599553 0.02821655 0.18778656 0.12817259 0.16854654

0.01031642 0.04990657 0.00041211 0.07939821 -0.00918845 0.00487251

-0.00212752 -0.00214860 -0.00248301 0.00337934 0.00170556 -0.00070350

Table 15.

Coefficients for Bole Length Equations

-

Non-Oaks

Cluster

BO BOLE LENGTH
=

B1 BO
+

B2
+

B3
+

B4
+

B5 B5*(PLCR) -22:33659577 6.98078503 -28.42 163995 -518.20492317 -185.78633161 -47. 16795576 -169. 77069085 -14.90279471 -98.80151689 29.83539272 -10.45425679 8.28936818 45.17757611 -0.88981247 -4.08671585 -40.20323175 -7.02088284 -4.09345350 - 18. 50935393 4.55908838

B1*(DBHT2)

B2*(SITE)

B3*(BALT2)

B4*(TOTA2)

+

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

11 .99805585 -26.78619372 11.17414893 218.49703327 127. 17798701 31.60793374 90.32717505 3.82422509 46.3 1748394 -32.53252352 -13.69411907 -10. 23490666 -23.04574406 10.05131164 8. 56475765 36.56004163 13.13006481 0523 1430 9. 7. 07211337 8.12592218

2.57071598 4.27470285 3.45302253 4.08658564 1.08279686 2.70940679 2.62105177 3.48634683 5.27646110 4.94947841 3.40863928 3.86878981 3.00995490 2.80981692 2.30095453 2.55226587 2.73827305 3.17035443 3.06648069 2.58756776

0.10239446 0.3 1258576 0. 06477151 0.07997534 0.18646014 0.03503347 0.15874811 0.09442930 -0.16147069 0. 03150258 0.20786327 0. 02961025 0.18539977 0. 122914 07 0.10073394 0. 02471245 0.06880837 0. 07022516 0.08620558 0.12749414

01745911 -0. 0.01320519 0. 00325131 0.01959178 -0. 15632181 -0. 00613777 -0.02722459 0.00060728 0.02861530 0.10328971 -0.00447535 0.01115720 -0. 071567 18 -0.01969680 -0.02510794 -0. 196641 02 -0. 00712913 -0.00520148 -0.00615299 0.01032247

-0.00026888 -0.00768636 -0.00099061 -0. 00893186 -0.00746105 -0.00276892 -0.00432662 0.00086728 0.00154920 0.00084069 0.00184714 0.00176952 0.00043494 -0.00053119 -0. 00050817 -0.00160741 -0.00050708 -0. 00196615 -0. 00117936 -0.00244715

lr\

~7 L~~L~?L~

AOn~~nlOAl

MULTIPURPOSE FOREST PROJECTION SYSTEM
MULTIPURPOSE FOREST PROJECTION SYSTEM

39

39

Table 16. Mortality Clusters

Pine

Oak
U131MPH BROA PROF5O BWOA ULWOA L131MPH P131MP V11OA POTHERA LROF4O LROF5O PWOA PROMPH

Non-oak

Cl uster 1. 2. 3.
P131MH P110F32 L131F31 P110F40 P132A L111F4O U111F22 P11OF5O BOTHERMPH V400A BSOFTA P611F50 U400A P611F31 L691A

4. 5.

P812F50 P802F50

P400A LSOFTA L316A PHARDA LROMPH UROMPH VWOA P611MPH USOFTMPH L200MPH U611F60 LHARDA L694MPH VOTHERA UROF6O L6L1MPH L653A U694A U611MPH LUPOTHERA L611F60 UDARDA USOFTF6O U611MPH L222F60

6. 7. 8. 9.

L121F21 L111F22 U121A B132A U1iQA LUOTHERA P110F31 VBOTHERA L121MPH L128A U131F31 L11MPH V132A V131A

LROF6O UROF5O L820F50 L820F60 U827F50

10.

RS VROA

11. 12.
13. 14. 15.

L621A U316A P621A L694F40 P31 6A L694F60 PSOFTA

Table 17. Coefficients for Mortality Equations - Pines

Cluster

BO

81

B2

B3

B4

85

B6

X = BO + B1*(TAL_T) DBHT1 > = 5.0 1 9.17194811 2 -0.78041095 3 8.53440500 4 5.80978614 5 13.16736467 7' 5.51188474 8 3.84942026 9 0.14503691 10 4.10658452 DBHT1 < 5.0 1 -4.82795624 2 -4.22323488 3 -6.10969109 4 -6.73120578 5 28.67379618 7' 1.06453275 8 -2.74390748 0.82525051 9 10 9.23707964

+ B2*(TO_TA1) + B3*(PLCR) + B4*(DBHT1) + B5*(SITE) + B6*(BAL_T1)

-0.00573346 -0.00466196 -0.00113172 -0.00100925 -0.01278102 -0.00436770 0.00044416 -0.00520571 0.00270978

-0.00016929 -0.00001475 0.00014907 0.00092069 0.00046394 0.00012803 -0.00002004 0.00028770 0.00185630

-13.4916 2.98230661 -11.8552 -8.25446 -24.9957 6.76269 -2.06575 0.97895478 -7.24557

-0.03196000 0.17228982 0.03491021 0.04367029 -0.02969901 0.02424753 0.14475845 -0.10301292 0.35603640

-0.00314735 0.00540931 -0.01246882 -0.01022889 -0.00161554 0.00636203 -0.01412170 0.03546311 -0.03159388

-0.02076203 0.01340063 -0.01759301 -0.01095206 -0.03760419 -0.00618504 -0.01325697 -0.00209673 -0.01583754

-0.00053386 0.00113204 -0.00040948 -0.00050135 -0.00228464 -0.00122383 -0.00091103 -0.00126769 -0.00190108

-0.00018473 -0.00060606 0.00039518 -0.00011996 0.00210310 0.00031542 -0.00009472 0.00019524 -0.00010767

12.9987 15.9906 14.3948 17.531 71.2498 9.761442 13.5042 3.491764 -3.73246

0.27407612 0.74043693 0.41438918 0.50903846 -0.16246518 0.27397747 0.40923931 0.39636065 0.57870286

0.00686293 -0.04388052 0.01152417 0.01406976 -0.01241976 -0.03263240 -0.01659749 -0.02809345 -0.11775905

-0.00560184 -0.01016036 -0.01153440 -0.00721299 0.03520507 -0.00748562 -0.00237869 -0.00566541 0.00110457

c C

m X m

Where, PMORT 1.0 = ----------------1.0 + exp(X)

z
-

"Cluster 6 combined with Cluster 7.

l--i

0 z

Table 18. Coefficients for Mortality Equations

-

Oaks

Cluster

B0

B1

B2

B3

B4

B5

X = INTERCEPT + B1*(TALT1) + B2*(TOTA1) + B3*(PLCR) + B4*(DBHT1) + B5*(SITE)
DBHTI >= 5.0

1 2 3 4 5
6

2.66936794 2.55396818 3.97679002 -48.18904781 10.01225107

7.39530461

-0.00258932 -0.00151891 -0.00356568 0.00208279 -0.00668616
-0.00142417

-0.00036896 -0.00034787 0.00012581 0.00139744 0.00085341
-0.00042258

-0.64893593 2.11597262 -2.07787545 87.08852827 4.05546436
-9.59791605

-0.06867221 -0.05036853 -0.06960713 -0.03961656 -0:08388281
-0.08285416

0.00814298 0.00446648 0.00108828 0.07226881 -0.08536477
0.01169516

7 8 9 10 DBHT1 1 2 3 4 5 6 7 8 9 10
<

12.63471990 2.17917154 15.35134699 4.81208046 5.0 -1.05072284 -2.07158146 -0.94729026 -12.46268269 -0.18656351 0.00440130 2.43853865 -4.12711915 -1.48608659 1.19924067 Where,

-0.00255722 -0.00819368 0.00480380 -0.00017567 -0.00149959 -0.00108736 -0.00199005 -0.00223315 -0.00162776 -0.00106971 -0.00112802 -0.00184693 -0.00177910 -0.00098588

-0.00033127 -0.00021723 -0.00005074 -0.00015810 0.00039873 0.00035022 0.00135980 0.00246556 0.00111065 0.00054342 0.00057549 0.00234250 0.00074931 0.00079832

-13.82405997 2.82336598 -33.41383007 -6.14094769 5.47045772
6.00604069

-0.15986485 -0.16705102 0.13909130 -0.04979528 0.08607928

-0.00262834 0.01518120 0.02858400 0.00237904 -0.01258862
-0.00504600

0.24231681

3.41293074 17.74409086 -0.60628083 0.28191582 -3.20578424 14.79079526 1.98237548 0.30136686

0.09162208 0.80306420 0.59732820 0.17824063 0.46917434 0.36799232 0.40730093 -0.03605216

-0.00310951 0.0101013 -0.00441235 0.00072588 -0.00927278 -0.04704037 -0.00067373 -0.01509243

1.0
PMORT
=------

1.0

+

exp(X)

Table 19. Coefficients for Mortality Equations - Non-Oaks Cluster BO B1 B2 B3 B4 B5

+ X - INTER ICEPT BI*(TAL TI) + B2*(TOTAI) + B3*(PLCR) + B4*(DBHT1) + B5*(SITE) DBHT1 >a 5.0 0.00010350 -3.08200601 -0.06434783 -0. 00251632 1 3.52714453 -0.00000318 0.37657831 0.03534203 0.00091330 2 1.98549664 -0.00098240 -8.58074263 -0.15716321 0.00291256 3 4.72968692 -0.00020498 38.48862230 -0.01891203 -0. 00265292 4 -16.81088533 -0.00002434 -0.70263276 -0.08255141 5 2.43497118 -0.00052694 -0.00021588 -6.31924092 -0.07992003 -0. 00351118 6 5.90065766 -0.00057050 3.99864451 -0.00297943 -0. 00281550 7 4.99455867 -0.00007706 -9.43457523 0.10318536 0.0051363 1 8 3.69710630 -0.00014222 -4.50631308 -0.10533885 9 -0. 00397938 4.10017365 0.00050530 -3.64710591 -0.11436720 00459870 -0. 10 1.93475993 -0.00045265 -75.06607936 -0.36238716 -0. 00239376 40.53076008 11 -0.00007703 -18.52463026 -0.11335788 00565973 -0. 12 9.47821152 -0.00001509 -47.37625847 0.13138042 0.01493787 13 20.58806257 0.00097740 -0.29686558 -0.06398131 14 2.49851174 -0. 00186261 -0.00054068 -7.77034764 -0. 14743483 -0. 00136685 7.15691228 15 DBHT1 < 5.0 0.00107019 1.12726578 0.255'11728 -0.00126790 1 -0.04951055 0.00029503 -0.60844971 0.23001919 00065551 -0. 2 0.86486933 0.00250079 2.96416697 0.15472986 -0. 00165916 3 -1.07471291 0.00111207 8.40329369 -0.01019835 -0.00195444 -1.69016462 4 0.00031428 0.99304544 0.15787323 -0. 00053517 5 0.22335370 0.00012862 2.99308524 0.37004638 -0. 00013888 6 -0.68939650 0.00084213 -12.79533900 0.03692028 00158603 -0. 10.97160998 7 -0.00063616 43.48443703 0.01971491 0.00018918 8 -14.55606230 0.00074221 0.96427222 0.09311962 -0.00093940 1.13206962 9 0.00038980 11.60855380 0.21544756 -0.00090392 -5.02564235 10 0.00085730 10.06941333 -0.18526511 -0.00220658 1.44205600 11 0.00133600 1.03715421 0.01621064 -0. 00221794 -2.12424916 12 0.00020831 -0.86208585 0.38325174 -0. 00065776 2.25726841 13 0.00085802 7.34648851 0.19918127 -0.00045306 -2.61098182 14 0.00041461 -13.06963919 0.27919167 -0.00084789 7.52002453 15 Where,
1 .0 PMORT
=---------A

0.01785795 -0.00358113 0:04763157 0.01586478 0.00429476 0.00799488 -0.02459980 0.01038003 0.02215372 0.02830042 0.00833487 0. 03130244 0.02200683 -0.00527276 0.01537900 -0.00032126 0.00115650 -0.00443655 -0. 00881217 0. 00016156 -0. 01016140 -0. 04966315 -0.05281810 -0.00853496 -0.00083931 -0.04363228 0.02848830 -0. 01896380 -0.00835737 -0.01480949

r-

C
I-

m x -v m m z

50

0

1.0 + exp(X)

z

MULTIPURPOSE

FOREST

PROJECTION

SYSTEM

43

APPENDIX 4. VOLUME EQUATIONS This appendix documents the publications from which the various volume equations were selected. Board foot equations were selected from the set presented by Parker (J). For the pines, the equation for Scribner log rule with a form class = 78 is used. The hardwoods use the equation for the Doyle log rule with a form class = 78. Sawtimber height is calculated for these equations, by utilizing a set of equations from the Southeastern Forest Survey Unit, which predicted sawlog length = f(dbh)'. The number of rough cords per tree is predicted using an equation presented by Merrifield and Foil (I). The equation selected estimates rough cords for form class 77. These equations were designed for estimation of Southern Pine pulpwood to a 3-inch top (d.o.b.). However, the projection system uses the equation for both pines and hardwoods. Bole length (which is to a 4-inch top) is used as the merchantable height in these equations. Therefore, caution should be exercised with these figures until more appropriate volume equations can be found. Considerably more information was available to predict both weight and cubic foot volumes. The equations selected are presented in table 1 and table 2. The capital letter in the table refers to the publication in the Literature Cited section which is being used for the species and region combination. A small s or h symbolizes that the equations for miscellaneous soft hardwoods or hard hardwoods are respectively being utilized. For those equations which require total height, a set of equations is again used from the Southeastern Forest Survey Unit, which predicts total height = f(dbh 2).
'Personal communications, December 1987, J.P. McClure, FIA, Southeastern Forest Experiment Station, Asheville, North Carolina, 28804.

44

44 ALABAMA AGRICULTURAL EXPERIMENT STATION
Table 1. Green Weight Equations for Wood and Bark of Stem to a 4-inch

Top Used inGA-TWIGS Species code Hard hardwoods 318 370 400 531 591 802 806 812 820 822 827 831 832 833 834 835 837 901 540 Soft hardwoods 221 316 460 611 621 693 694 731 950 Pines 110 111 121 129 131 132 260
I

Lcp, Ucp Eh A 0 E A 0 E C C E K E N C 0 G B 0 0 L Es F E M E E E C 0 F F F F F F F

Inn Ii~an ~n I-n-iwii

Pie

Val

Blu Oh A 0 0 A 0 0 B B E K E N C 0 G B 0 0 L Ds F 0 M B 0 0 0 C 0 F F F F F F F

Equivalent species 311, 491, 591, 680

Ch A D C A 0 C C C E K E N C 0 G B 0 0 L Cs F C M C C 0 E C 0 F F F F F F F

Bh A 0 B A 0 B B B E K E N C 0 G B 0 0 L Bs F 0 M B B 0 0 C 0 F F F F F F F

601, 602

813 804, 823, 825 826

830, 828, 838 807, 816, 819, 824 840, 841, 899

313, 555, 651, 652, 653, 691, 740, 920, 970 220

MULTIPURPOSE FOREST PROJECTION SYSTEM Table 2. Cubic Foot Equations for Wood Only of Stem to a 4-inch Top Used in GA-TWIGS Species code Hard hardwoods 370 400 540 802 806 812 820 822 827 831 832 833 835 837 901 Soft hardwoods 221 316 460 611 621 691 693 694 731 950 970 Pines 110 111 121 129 131 132 260

45

Lcp, Ucp Eh 0 E L E C C E K E N C D 0 B 0 D Es F E M E E E 0 E C D C F F F F F F F

Pie Ch D 0 C L C C C E K E N C D 0 B D D0 Cs F C M C C E D E C 0 C F F F F F F F

Val Bh D 0 B L B B B E K E N C D 0 B D 0 Bs F D M B B E D 0 C 0 C F F F F F F F

Blu Dh D D L D B B E K E N C D B D D D s F D M B D E 0 0 C 0 C F F F F F F F

Equivalent species 311, 318 491, 531, 591, 680 601, 602

813, 828, 830, 834 804, 823, 825 826

807, 816, 819, 824, 840, 841, 899

313, 554, 651, 652, 653, 691, 740, 762, 920, 970 220

I46
T V

ALABAMA AGRICULTURAL EXPERIMENT STATION

Literature Cited
(A)
BRENNEMAN, B.R. AND R.F. DANIELS. 1982. A Case Study, with Alter-

(B) (C)
(D)

(E)

natives, for Calculating Biomass Yield. In: Baldwin, V.C.; Lohrey, R.E., eds. - Proc. of 1982 Southern Forest Biomass Working Group Workshop, June 16-18, 1982, Alexandria, La. Louisiana State University. Baton Rouge, La. pp. 105-110. . 1986. Weight and Volume Equations for Southeastern Tree Species [unpublished office report]. Athens, Ga. U.S.D.A. For. Ser. S.E. For. Exp. Sta. . 1986. Total Tree Weight, Stem Weight, and Volume Tables for Hardwood Species in the Southeast. Georgia Forestry Commission. Ga. For. Res. Pap. 60. CLARK, A., III, D.R. PHILLIPS, AND D.J. FREDRICK. 1986. Weight, Volume, and Physical Properties of Major Hardwood Species in the Upland-South. U.S. For. Ser. S.E. For. Exp. Sta. Res. Pap. SE-257. . 1986. Weight, Volume, and Physical Properties of Major Hardwood Species in the Piedmont. U.S. For. Ser. S.E. For. Exp. Sta. Res. Pap. SE-255.
. 1986. Weight, Volume, and Physical

(F)

Properties of Major Hardwood Species in the Gulf and Atlantic Coastal Plains. U.S. For. Ser. S.E. For. Exp. Sta. Res. Pap. SE-250. (G) AND J.G. SCHROEDER. 1986. Weight, Volume, and Physical Properties of Major Hardwood Species in the Southern Appalachian Mountains. U.S. For. Ser. S.E. For. Exp. Sta. Res. Pap. SE-253. (H) MCNAB, W.H. AND A. CLARK, III. 1982. Total Tree and Major Component Green Weight of White Pine and Hemlock in North Georgia. Georgia Forestry Commission. Ga. For. Res. Pap. 31. (I) MERRIFIELD, R.G. AND R.R. FOIL. 1967. Volume Equations for Southern Pine Pulpwood. Louisiana State University, Agricultural Experiment Station, Hill Farm Experiment Station, Homer, La. Hill Farm Facts, Forestry 7. (J) PARKER, R.C. 1972. Regression Equations for the Mesavage and Girard Form-class Volume Tables. Virginia Polytechnic Institute and State University, Blacksburg, Va., Cooperative Extension Service, Pub. 501.
(K)
.

1981. Willow Oak Volume and Weight Tables for the

(L)
(M) (N)

Mississippi Delta. USDA For. Ser. So. For. Exp. Sta. Res. Pap. SO-173. SCHLAEGEL, B.E. 1984. Overcup Oak Volume and Weight Tables. USDA For. Ser. So. For. Exp. Sta. Res. Pap. SO-207.
. .

1984. Green Ash Volume and Weight Tables. USDA For. 1984. Sugarberry Volume and Weight Tables. USDA For.

Ser. So. For. Exp. Sta. Res. Pap. SO-206. Ser. So. For. Exp. Sta. Res. Pap. SO-205.

MULTIPURPOSE

MULIPUPOSE

FOREST

FOREST PROJECTION SYSTEM

PROJECTION

SYSTEM

47

4

APPENDIX 5. SPECIES CODE, COMMERCIAL TREES Yellow Pines
107 110 111 115 121 123 126 128 131 132 Sand pine Shortleaf pine Slash pine Spruce pine Longleaf pine Table-Mt. pine Pitch pine Pond pine Loblolly pine Virginia pine

Pinus clausa Pinus echinata Pinus elliottii Pinus glabra Pinuspalustris Pinuspungens Pinus rigida Pinus serotina Pinus taeda Pinus virginiana

Other Softwoods
010 043 060 090 129 221 222 241 260 Fraser fir Atlantic white-cedar Eastern redcedar Red spruce White pine Baldcypress Pondcypress Northern white-cedar Eastern hemlock

A bies fraseri Chamaecyparis thyoides Juniperus virginiana Picea rubens Pinus strobus Taxodium distichum var. distichum Taxodium distichum var. nutans Thuja occidentalis Tsuga canadensis

Soft Hardwoods
313 316 317 330 460 555 580 601 611 621 651 652 653 691 693 694 731 740 762 920 950 970 Boxelder Red maple Silver maple Buckeye Hackberry Loblolly-bay Silverbell (in mts. ) Butternut Sweetgum Yellow-poplar Cucumbertree Magnolia Sweetbay Water tupelo Blackgum (upland) Blackgum (lowland) American sycamore Cottonwood Black cherry Willow American basswood Elm

Acer negundo A cer rubrum Acer saccharinum Aesculus spp. Celtis occidentalis Gordonia lasianthus Halesia spp. Juglans cinerea Liquidambarstyraciflua Liriodendron tulipifera Magnolia acuminata Magnolia spp. Magnolia virginiana Nyssa aquatica
Nyssa

sylvatica

Nyssa sylvatica Platanus occidentalis Populus spp. Prunus serotina Salix spp. Ti/ia americana
Ulmus spp.

48

48 ALABAMA AGRICULTURAL EXPERIMENT STATION

Hard Hardwoods
311 318 370 371 400 491 521 531 540 552 591 602 680 802 804 806 812 813 817 820 822 823 825 826 827 830 831 Florida maple Sugar maple Birch (except yellow) Yellow birch Hickory Flowering dogwood Persimmon (forest grown) American beech Ash Honeylocust American holly Black walnut Red mulberry White oak Swamp white oak Scarlet oak Southern red oak Cherrybark oak Shingle oak Laurel oak Overcup oak Bur oak Swamp chestnut oak Chinkapin oak Water oak Pin oak Willow oak Chestnut oak Northern red oak Shumard oak Post oak Black oak Live oak Black locust Acer barbatum Acer saccharum Betula spp. Betula alleghaniensis Carya spp. Cornus florida Diospyros virginiana Fagus grandifolia Fraxinus spp. Gleditsia triacanthos ilex opaca Juglans nigra Morus rubra Quercus a/ba Quercus bicolor Quercus coccinea Quercus falcata Quercus falcata var. pagodaefolia Quercus imbricaria Quercus laurifolia Quercus lyrata Quercus macrocarpa Quercus michauxii Quercus muehlenbergii
Quercus
Quercus

palustris

nigra

832
833 834 835 837 838 901

Quercus phellos Quercus prinus Quercus rubra Quercus shumardii Quercus stellata Quercus velutina Quercus virginiana Robinia pseudoacacia

MULTIPURPOSE FOREST PROJECTION SYSTEM

49

4

Miscellaneous Species
310 315 319 341 352 391 421 451 471 521 548 581 641 660 661 692 701 711 712 721 722 760 807 816 819 824 840 841 899 931 .999 Chalk maple Striped maple Mountain maple Ailanthus Serviceberry Blue beech American chestnut Catalpa Eastern'redbud Persimmon (field grown) American mt. ash Carolina silverbell (except mts.) .Osage-orange Domestic fruit (apple etc.) Chinaberry Ogeechee gum Eastern hophornbeam Sourwood Royal paulownia Redbay Planer-tree (water elm) Fire cherry Bluejack oak Bear oak Turkey oak Blackjack oak Dwarf post oak Dwarf live oak Other scrub oaks Sassafras Other miscellaneous trees Acer saccharum var. leucoderme Acer pensylvanicum Acer spicatum Ailanthus spp. Amelanchier spp. Carpinus caroliniana Castanea dentata Catalpa spp. Cercis canadensis Diospyros virginiana Pyrus americana Ha/esia carolina Mac/ura pomifera Ma/us spp. Me/ia azedarach Nyssa ogeche Ostrya virginiana Oxydendrum arboreum Paulownia tomentosa Persea borbonia Planera aquatica Prunus pennsylvanica Quercus incana Quercus ilicifolia Quercus laevis Quercus marilandica Quercus stellata spp. Quercus virginiana spp. Quercus spp. Sassafras albidum

50

ALABAMA

AGRICULTURAL

EXPERIMENT

STATION

APPENDIX 6. FOREST TYPE DEFINITIONS White Pine - Hemlock (Code 4) - Forests in which eastern white pine and hemlock, singly or in combination, comprise a majority of the stocking. Loblolly Pine Plantation (Code 5) - Forests in which loblolly pine was artificially regenerated with acceptable survival and comprises a plurality of the stocking. Shortleaf Pine Plantation (Code 6) - Forests in which shortleaf pine was artificially regenerated with acceptable survival and comprises a plurality of the stocking. Longleaf Pine Plantation (Code 7) - Forests in which longleaf pine was artificially regenerated with acceptable survival and comprises a plurality of the stocking. Longleaf Pine (Code 21) - Forests in which southern yellow pines, singly or in combination, comprise a plurality of the stocking, and in which longleaf pine contributes the most stocking of the pines. Slash Pine (Code 22) - Forests in which southern yellow pines, singly or in combination, comprise a plurality of the stocking, and in which slash pine contributes the most stocking of the pines. Loblolly Pine (Code 31) - Forests in which southern yellow pines, singly or in combination, comprise a plurality of the stocking, and in which loblolly pine contributes the most stocking of the pines. ShortleafPine(Code 32) - Forests in which southern yellow pines, singly or in combination, comprise a plurality of the stocking, and in which shortleaf pine contributes the most stocking of the pines. VirginiaPine (Code 33) - Forests in which southern yellow pines, singly or in combination, comprise a plurality of the stocking, and in which virginia pine contributes the most stocking of the pines. Redcedar (Code 35) - Forests in which redcedar comprises a plurality of the stocking. Pond Pine (Code 36) - Forests in which southern yellow pines, singly or in combination, comprise a plurality of the stocking, and in which pond pine contributes the most stocking of the pines. Pitch Pine (Code 38) - Forests in which southern yellow pines, singly or in combination, comprise a plurality of the stocking, and in which pitch pine contributes the most stocking of the pines. Oak-Pine (Code 40) - Forests in which hardwoods (usually upland oaks) comprise a plurality of the stocking but in which pines com-

MULTIPURPOSE

MULTIPURPOSE

FOREST

FOREST PROJECTION SYSTEM

PROJECTION

SYSTEM

51

5

prise 25 to 50 percent of the stocking. (Common associates include gum, hickory, and yellow-poplar.) Oak-Hickory (Code 50) - Forests in which upland oaks or hickory, singly or in combination, comprise a plurality of the stocking, except where pines comprise 25 to 50 percent, in which case the stand would be classified oak-pine. (Common associates include yellowpoplar, elm, maple, and black walnut.) Chestnut Oak (Code 52) - Forests in which chestnut oak (Quercus prinus) comprises a plurality of the stocking. Southern Scrub Oak (Code 57) - Forests in which blackjack, bluejack, turkey, dwarf post, and bear oak, singly or in combination, comprise a plurality of the stocking. Oak-Gum-Cypress (Code 60) - Bottomland forests in which tupelo, blackgum, sweetgum, oaks, or southern cypress, singly or in combination, comprise a plurality of the stocking, except where pines comprise 25 to 50 percent, in which case the stand would be classified oak-pine. (Common associates include cottonwood, willow, ash, elm, hackberry, and maple.) Elm-Ash-Cottonwood (Code 70) - Forests in which elm, ash, or cottonwood, singly or in combination, comprise a plurality of the stocking. (Common associates include willow, sycamore, beech, and maple.)

With an agricultural rcearch unit in everv maeccssoil area, Auburn niversitV serves e needs of tield crop, livestock, forestry, and horticultural producers in each region in Alabama. Ever, citizen of the State has a stake in this research program, since any advantage from new and more economical way's of producing and handling farm products directly benefits the consuming public.

7

8

0 02 06 0O 0

0

0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

*j E. V. Smith Research Center, Shorter.

Main Agricultural Experiment Station, Auburn.

Tennessee Valley Substation Belle Mina Sand Mountain Substation, Crossville North Alabama Horticulture Substation Cullman Upper Coastal Plain Substation, Winfield Forestry Unit, Fayette County Chilton Area Horticulture Substation. Clanton Forestry Unit, Coosa County Piedmont Substation, Camp Hili Plant Breeding Unit, Talassee Forestry Unit. Autauga County Prattville Experiment Field. Prattville Black Belt Substation, Marion Junction The Turn pseed-Ikenberry Place. Union Springs Lower Coastal Plain Substation. Camden Forestry Unit. Barbour County Monroeville Experiment Field, Monroeville Wregrass Substation, Headland Brewton Experiment Field. Brewton Soion Dixon Forestry Education Center. Covington ano Escambia counties 20 Ornamental horticulture Substation Spring Hill 21 Gulf Coast Substation. Fairhope