ASPEN - A Circumboreal Growth and Yield Model for Populus tremuloides and
P. tremula: User's Guide
George E. Host(1) and Donald A. Perala(2)
ASPEN is an empirical simulation model that projects the growth and yield of aspen (Populus
tremuloides and P. tremula) stands from establishment to breakup. The model incorporates the
system of equations developed from growth and yield data from throughout the range of P.
tremuloides and P. tremula; the model should thus be applicable throughout the circumboreal
region. The model runs on an annual time step and predicts total yields in number of trees, basal
area, and biomass, as well as merchantable yields in cubic feet and cords for user-specified
utilization standards, and in Scribner board feet. Outputs are in the form of stand tables and stocking guides. The model supports silvicultural operations such as thinning, as well as management for multiple products.
ASPEN is a computer simulation model that projects the growth of aspen (Populus tremuloides,
P. tremula) stands from establishment to breakup. The model incorporates the system of
equations presented in Perala et al. 1996. Data for these equations were compiled from
throughout the range of P. tremuloides and P. tremula; the model should thus be applicable
throughout the circumboreal region. The ASPEN model was originally developed to provide
growth and yield information to the Forest Management Advisory System (FMAS), a decision
support system for managing aspen and red pine (Nute et al. 1995, Rauscher et al. 1995).
ASPEN can also be run as a stand-alone model, with or without a user interface. The model runs
on an annual time step and predicts total yields in number of trees, basal area, and biomass, as
well as merchantable yields in cubic feet and cords for user-specified utilization standards, and in
Scribner board feet. The model supports silvicultural operations such as thinning, and supports
multiproduct management (e.g. pulp, sawtimber, special products such as cabin timbers).
The model can be run in batch or interactive mode. In batch mode, the user simply supplies an
input file that describes the stand condition and objectives, and an updated output file is returned.
This output file can serve as an input to a subsequent run. In interactive mode, the user has the
option of altering stand conditions, product objectives or stopping rules (e.g. stop at culmination
of MAI or need for thinning). Model outputs in interactive mode include a stand table and
stocking chart, as well as the output files required for subsequent runs.
The input file contains data that defines the stand and supplies management objectives and stopping conditions (Table 1). By default, this file is called PASS2GRO.DAT, although the user has the option of specifying a different file for input (see below). The variables are read from an ASCII file in space or comma-delimited format.
|Table 1. Input variables for PASS2GRO.DAT|
|Trees per Acre||# trees/ac||Required|
|Diameter at Breast Height||in||Required|
|Site Index||height at age 50 (ft)||Required|
|Mean July Temperature||oF||Required|
|Top Diameter for Pulp||in||Required|
|Top Diameter for Special Products||in||Required|
|Stopping Rule||integer 1-10||Required|
|Residual Trees||# trees/ac||Passed|
|First Thinning||0, 1||Required|
All nineteen variables must be present in the input file. Those that are 'required' are used by the model. Variables that are 'passed' are calculated by the model, and returned in the output file.
Basal area, for example is calculated as a function of number of trees and mean stand dbh.
The initial stand conditions are described by the first six variables: Age, Current Year, Basal Area, Trees per Acre, Stand Quadratic Mean Diameter, and Dominant Height. The next two variables, Site and Mean July Temperature, adjust the model for site and regional growing conditions, as described in Perala et al. in press. Minimum top diameters for pulp and special products are specified variables 9 and 10, respectively. The Stopping Rule and Stopping Value define the conditions for terminating the current run. In addition, Stopping Rules less than 6 will trigger a thinning prior to the model run. The Stopping Rules consist of integer values from one to ten, as defined in Table 2.
|Table 2. Rules, conditions and values for triggering initial thinning and terminating runs.|
|Stopping Rule||Stopping Condition||Stopping Value||Thin before Run?|
|1||Stop if thinning is needed to achieve a product objective||Number of trees to remain after thinning||Yes|
|2||Stop at culmination of mean annual increment||None needed; MAI is calculated based on product objective||Yes|
|3||Stop when growth reaches a specified dbh||Target dbh (in)||Yes|
|4||Stop at user-specified number of years||Number of years||Yes|
|5||Stop at user-specified rotation age||Rotation Age
(stand age in years)
|6||Stop if thinning is needed to achieve a product objective||Number of trees to thin to
|7||Stop at culmination of mean annual increment||None needed; MAI is calculated based on product objective||No|
|8||Stop when growth reaches a specified dbh||Target dbh (in)||No|
|9||Stop at user-specified number of years||Number of years||No|
|10||Stop at user-specified rotation age||Rotation Age
(stand age in years)
The variables Residual Trees, Harvested Wood, and Rotation Age, are not required by the model,
but are used for interfacing with the Forest Management Advisory System (FMAS) developed by
Rauscher et al. (1995). Residual Trees stores the growth of trees other than aspen, as predicted by
FMAS. Harvested Wood is a cumulative sum of timber removed from all thinning operations, in
cubic feet. Rotation Age is also predicted by the FMAS. If stopping rules 5 or 10 are selected,
the user can override the value in Rotation Age by setting a new Stopping Value.
The next variables are binary variables used in model equations. The First Thinning variable is
used in equation  of Perala et al. in press; the appropriate model differs if a stand has been
thinned previously. Similarly, the binary variable Tremula indicates whether the stand is
composed of P. tremula (Tremula=1) or P. tremuloides (Tremula=0; Perala et al. in press).
The variable Goal sets a user-specified product objective, as follows:
0 - manage for biomass production (no specified top diameter)
1 - manage for pulp production (user-specified top diameter)
2 - manage for sawtimber (top diameter = 6 in)
3 - manage for special products (user-specified top diameter)
The final variable, WhoDat, is a flag that determines whether the model runs in interactive or
batch mode. A WhoDat value of 1 loads the user interface, while a WhoDat value of 0 runs the
model in batch mode.
The following shows a valid PASS2GRO.DAT file:
90 2064 4.5 3.9 14.6 89.4
70 60 4 9
1 0 738 131 200
0 0 3 1
The general flow of the program is shown in Figure 1. An input file containing initial stand
information and program control data is read and evaluated. The WhoDat variable in the input
file determines if the program is to run in batch or interactive mode. Initial volumes and residual
trees are calculated, and the stand is thinned, if required. The program then enters an annual loop,
in which diameter growth, survival, and volume increments for that year are calculated. The
appropriate Stopping Rule is then evaluated to determine if the run should be terminated. Upon
termination, the input file is updated with new values, and updated stocking charts and stand
tables may be viewed.
Installing and Running the Program
The ASPEN program and data files are delivered in compressed format, in the file ASPEN.ZIP.
These can be installed by copying this file to an appropriate directory on a hard disk (e.g.
c:\ASPEN), making this the default directory (CD \ASPEN) and typing PKUNZIP ASPEN.
The program is started by typing ASPEN in the default directory. The user can use a different input file by typing that filename on the command line, e.g.
will use myfile.dat as the input file.
A simple user interface allows interactive manipulation of the input and output variables. The
required variables read in from the PASS2GRO.DAT file are presented on the data input screen
shown in Figure 2. The arrow keys allow the user to move from field to field. In addition, the
<ENTER> and <ESC> keys allow the user to move forward and backward, respectively. Fields
are checked on data entry to protect against out-of-range values. The F1 key is used to run the
model when data entry is complete.
A run-time screen shows the year-by-year progress of the model, displaying age, basal area,
number of trees per acre, quadratic mean diameter, and height (Figure 3). When the run is
complete (i.e. one of the Stopping Rules have been met), the user presses a key to display an
output menu. The user has the options of viewing the either the stand table or stocking guide,
running the model again, or exiting the program. The stand table (Figure 4) displays the starting
and ending stand conditions, as well as volumes of pulp, sawtimber and special products. The
user can toggle to a screen that displays biomass values (Figure 5), or return to the main output
The stocking guide displays the trajectory of the stand based on changes in the number of trees
and stand basal area (Figure 6). Points are written to the graph in five year intervals. The guide
shows the A, B, and C lines of traditional stocking charts, as well as the quadratic mean diameter
of the stand. Figure 6 shows a series of three consecutive model runs, with thinning operations at
the beginning of each run.
Option 3 allows the user to run the model again. The user is returned to the data input screen,
where variables or goals can be revised. Option 4 closes all files and returns the user to the
operating system. Upon exiting the model, the data in PASS2GRO.DAT is updated.
ASPEN was written and compiled using Microsoft QuickBASIC version 4, and runs under
MS-DOS, or in a Windows 3.1 or Windows95 MS-DOS box. The program runs under
conventional memory, and requires under a megabyte of disk storage. An EGA or VGA card is
required to display the stocking chart.
The ASPEN model is available here:
You can contact the senior author at
George E. Host, Natural Resources Research Institute, 5013 Miller Trunk Hwy., Duluth, MN 55811
or via the Internet: email@example.com
The U.S. Department of Agriculture or University of Minnesota cannot assure the accuracy,
completeness, reliability, or suitability for any purpose other than that reported. The recipient
may not assert any proprietary rights thereto nor represent it to anyone as other than a
Government-produced computer program.
The use of trade, firm or corporation names in this paper is for the information and convenience of the reader. Such use does not constitute official endorsement or approval by the U.S. Department of Agriculture or University of Minnesota of any product or service to the exclusion of others that may be suitable.
Nute, D. E., H. M. Rauscher, D. A. Perala, G. Zhu, Y. Chang, and G. E. Host. 1995. A toolkit
approach to developing forest management advisory systems in PROLOG. AI Applications in
Natural Resources 9:39-58.
Perala, D. A., G. E. Host, J. K. Jordan, and C. J. Cieszewski. A multiproduct growth and yield
model for the circumboreal aspens. Northern Journal of Applied Forestry 13(4):164-170.
Rauscher, H. M., D. E. Nute, D. A. Perala, G. Zhu, Y. Chang, and G. E. Host. 1995. The Forest Management Advisory System. AI Applications in Natural Resources. 9:60.
1.Research Associate, Natural Resources Research Institute , University of Minnesota, Duluth, MN 55811
2.Retired, Former Principal Silviculturist, USDA Forest Service, North Central Forest Experiment Station, Grand Rapids, MN 55744