Boxelder
Page 1.
Introduction: discussion of classification, seral assignment and monitoring.
Page 2.
Graph.
Page 3.
Instructions: plot setup and data collection using the Excel spreadsheet.
Page 4.
References.
The Boxelder classification and monitoring system was developed using tree and shrub
variables only. Herbaceous understory data were not collected for the Boxelder
ecological type. Resource managers need to consider the herbaceous component when
evaluating the overall ecological system.
Caution
Boxelder.xls is an Excel workbook containing the datasheet for field data collection. Copy the Excel file to a
PDA. The .xls file will be converted to a Pocket Excel file with a .pxl extension. The file contains a
spreadsheet with formulas that provide calculations necessary for classification of a site, using the variables
defined for this ecological type.
Be careful with the spreadsheet when loaded on a PDA. Although the formulas and cells are protected on the
PC version of Excel, protection of spreadsheets/worksheets is not an option on the Pocket version of Excel. It
is possible to erase formulas and stop the classification system. Be sure to make changes only within the empty
cells, and not in any of the calculation or formula cells such as cov%, probabilities, etc.
In addition, always:
•
Keep a backup copy of the original ‘Boxelder.pxl’ file on your PDA and,
•
Save the file under a new name after finishing data collection on each site, e.g. ‘Boxelder site1.pxl’,
‘Boxelder site2.pxl’…….. to avoid overwriting previous files with data.
Monitoring Seral Stages in Boxelder-Hackberry-Snowberry Ecological Type
Classification, seral assignment, and monitoring of Boxelder-Hackberry-Snowberry ecological type seral
stages is according to the method of Uresk (1990). Statistical analysis of field-collected data indicates the
presence of four seral stages: early, early intermediate (EINT), late intermediate (LINT) and late. A site is
classified, and assigned to a seral stage by entering a set of key plant measurements into the set of canonical
discriminant functions developed for this ecological type and estimating posterior probability for each seral
stage (SPSS, 1995). The site monitored is assigned to the stage with the highest posterior probability (See
Boxelder.xls program for methods).
For the Boxelder-Hackberry-Snowberry ecological type the key variables are the basal area of trees in
square feet per acre for Acer negundo (ACNE) and Celtis occidentalis (CEOC), and the percent canopy cover
of Symphoricarpos occidentalis (SYOC), using Daubenmire’s (1959) canopy cover method.
The seral stage assigned to a set of plant measurements is always associated with the greatest probability
value. If that value is 1, then succession is at the middle of the assigned seral stage. If the greatest probability
value is less than 1, then the second greatest probability value is used to determine the successional status
relative to middle of the assigned stage. For instance, if the second greatest probability is associated with an
earlier seral stage, succession is at the lower end of the assigned seral stage. On the other hand, if the second
greatest probability is associated with a later seral stage, succession is at the upper end of the assigned stage
(see example below).
When permanent macroplots are established to assess the vegetational trend of a site, re-sampling and
comparing the previously assigned seral stage to the current stage provides information about succession or
retrogression. On a finer scale, within-stage changes in vegetation successional status can be quantified by
comparing the probability values. For example, first-year field measurements collected for basal area of the
two tree species and canopy cover of snowberry on a site resulted in the following values: ACNE=25,
CEOC=31, and SYOC=17. Four years later the values obtained were: ACNE=27, CEOC=32, and SYOC=15.
Calculating the probabilities, we find:
ACNE CEOC SYOC
LATE LINT
EINT
EARLY
Stage
Year1
25
31
17
.005
.780
.141
.075
LINT
Year5
27
32
15
.003
.924
.033
.004
LINT
Difference
In both years, the site was classified and
assigned to the late intermediate seral stage; the
greatest probability values (.780 in Year1 and
.924 in Year5) are associated with LINT.
Placement within the late intermediate seral
stage is at the lower end of the stage, as
indicated by the second greatest probability
values (.141 for Year1 and .033 for Year5)
which are associated with an earlier seral stage
(EINT).
The site shows an upward
successional change of about .14, which is the
difference between the probabilities in Year1
and Year5.
The figure to the right illustrates this example.
These quantitative changes in the probabilities
can be used to evaluate and monitor the effects
of management alternatives.
.144
Late
Early Interm ediate
Late Intermediate
.78
Early
Year 1
1.0
.5
Late
1.0
.5
1.0
Early Intermediate
Late Intermediate
.92
1.0
.5
Early
Year 5
1.0
.5
Late
1.0
.5
1.0
Early Intermediate
Late Intermediate
.14
1.0
.5
Early
Change
1.0
.5
1.0
.5
1.0
.5
1.0
Page 1 of 4
Monitoring Seral Stages in Boxelder-Hackberry-Snowberry Ecological Type
Boxelder-Hackberry-Snowberry
Ecological Type by Seral Stage
70
Boxelder Basal Area (ft2 /acre)
Hackberry Basal Area (ft2 /acre)
Snowberry (%cover)
60
Mean values
50
40
30
20
10
0
LATE
L. INTERMEDIATE E.INTERMEDIATE
EARLY
Mean Values
Boxelder BA (ft2/a)
Hackberry BA (ft2/a)
Snowberry cover (%)
58
8
6
29
47
7
18
11
33
5
4
3
Mean values of plant variables through four stages of succession in a Boxelder-Hackberry-Snowberry
ecological type. Smoothed lines connect the means for each key plant variable at each stage. Graph
provides a guide for an approximate mixture of species variables at each seral stage.
Page 2 of 4
Monitoring Seral Stages in Boxelder-Hackberry-Snowberry Ecological Type
DATA COLLECTION,
SUMMARIZATION
30m
OR
AND
40m
CALCULATIONS
10m
40m
20m
Boxelder Overstory
Basal
CEOC
DBH
BA
5.1
20
9.2
66
5.4
23
6.7
35
6.3
31
12
113
4.7
17
15.8
196
3.3
9
7.9
49
8.4
55
11.6
106
9.4
69
11.4
102
5.3
22
1.5
2
ACNE
27
30m
30m
VEG TYPE LIMITS
ACNE
DBH
BA
4.5
16
8.9
62
2.3
4
2
3
1.1
1
3.2
8
10
79
11.7
107
5.6
25
4.4
15
5.8
26
6.8
36
3.5
10
6
28
13.8
149
2.9
7
2.7
6
2
3
3.1
8
2.3
4
1.9
3
10.7
90
4.6
17
2.4
5
7.1
40
3.6
10
40m
20m
30m
Enter Daubenmire classes 1 through 6
Xsect
CEOC
32
Frm
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Mean cov%
SYOC
3
1
4
1
1
2
5
6
4
1
1
5
3
1
1
4
1
3
3
1
4
1
2
2
5
2
TRANSECT
1. Within the Boxelder-Hackberry-Snowberry type, establish
a permanent plot with an area of 800m2. Ideal plot
dimensions are 20m x 40m; however, on some sites it may
be necessary to use two 10m x 40m plots. Along the plot
boundaries, establish two 30m transects, a minimum of
20m apart. (See figure above)
Recommendations are to monitor two sites per section
(640 acres) within the ecological type. Sites should be at
least ¼ mile apart.
2. Using the ‘Boxelder.xls’ spreadsheet (see figure at left)
installed on your PDA:
Record the DBH in inches of each Acer negundo (ACNE)
and Celtis occidentalis (CEOC) tree 1” or greater diameter
within the 800m plot.
Record canopy cover of Symphoricarpos occidentalis
(SYOC) within a 20cm x 50cm quadrat at one-meter
intervals along each transect (30 readings/transect) using
Daubenmire (1959) class codes 1 through 6.
Class Code
1
2
3
4
5
6
Cover Class
>0 - 5%
5 - 25%
25 - 50%
50 - 75%
75 - 95%
95 – 100%
Mid Point
2.5%
15%
37.5%
62.5%
85%
97.5%
The spreadsheet calculates basal area for each tree and for the
site, and mean percent canopy cover for snowberry for the
site. It also provides classification probabilities by seral stage
and the seral stage assignment.
‘Boxelder.xls’ Calculations
2
2
Basal area per tree (in inches) = DBH2 / 4 x 3.14.
1
SYOC
15
Seral Stage Classification Posterior Probability
LATE
LINT EINT ERLY
0.00
0.92 0.03
0.05
Assigned Stage
LATE INTERMEDIATE
PLOT LIMITS
BA (ft2 / acre) = Sum basal of all trees on plot, divided by 144 to
convert basal in inches to basal in feet. BA (ft) x 5.06 converts BA
per 800m2 to BA per acre.
Cover (%) = Sum of Daubenmire cover class midpoint values for
Snowberry divided by 60.
Page 3 of 4
Monitoring Seral Stages in Boxelder-Hackberry-Snowberry Ecological Type
REFERENCES
Daubenmire, R. 1959. A canopy-coverage method of vegetational analysis. Northwest Science.
33(1):43-64.
SPSS 12.0 for Windows. 1989-2003. SPSS Inc. 444 N Michigan, Ave, Chicago, IL 60611.
Uresk, D. W. 1990. Using multivariate techniques to quantitatively estimate ecological stages in a
mixed grass prairie. J. Range Mgt. 43(4):282-285.
Page 4 of 4