INFLUENCE OF ALTERNATIVE VEGETATION MANAGEMENT TREATMENTS ON PLANT COMMUNITY ATTRIBUTES: ABUNDANCE, SPECIES DIVERSITY,

AND STRUCTURAL DIVERSITY.

by

Pontus Mauritz Fredrik Lindgren

B.Sc. (Forestry), The University of British Columbia, 1995

A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE

in

THE FACULTY OF GRADUATE STUDIES Department of Forest Sciences

Faculty of Forestry

We accept this thesis as conforming to the

required standard

THE UNIVERSITY OF BRITISH COLUMBIA December 1999

© Pontus Mauritz Fredrik Lindgren, 1999

In presenting t h i s t h e s i s i n p a r t i a l f u l f i l m e n t of the requirements f o r an advanced degree at the U n i v e r s i t y of B r i t i s h Columbia, I agree that the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r reference and study. I f u r t h e r agree that permission f o r extensive copying of t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the head of my department or by h i s or her r e p r e s e n t a t i v e s . I t i s understood that copying or p u b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l gain s h a l l not be allowed without my w r i t t e n permission.

Department of

The U n i v e r s i t y of B r i t i s h Columbia Vancouver, Canada

Abstract

This study was designed to test the hypothesis that alternative vegetation management

treatments (manual cutting and cut-stump applications of glyphosate herbicide), applied

to young plantations, would decrease plant community abundance (crown volume index),

species diversity, and structural diversity. The experimental design consisted of nine

operational-sized plantations stratified on the basis of location and elevation into 3 blocks

(i.e., 1 control, 1 manual, and 1 cut-stump plantation per block), and with 5 permanent

strip-transects to sample vegetation within each plantation. Vegetation management

treatments did not significantly (P > 0.10) affect the crown volume index of the herb,

shrub, or coniferous tree layers. However, both manual and cut-stump treatments

significantly reduced the crown volume index of deciduous trees in the first post-

treatment year (P = 0.05 and P < 0.01, respectively). Due to prolific growth of stump

sprouts, the manual treatment effect did not extend past the first post-treatment year. In

contrast, the cut-stump treatment impeded sprouting and, relative to control and manual

treatments, significantly suppressed deciduous growth for at least four years (P < 0.05).

Species richness, diversity, and turnover of the herb, shrub, and tree layers were not

significantly (P > 0.10) different between treatments and control. Similarly, the structural

diversity of herb, shrub, and tree layers were also not significantly (P > 0.10) different

between treatments and control. By opening the canopy and decreasing the dominance of

the deciduous tree layer, both manual and cut-stump treatments showed greater total

structural diversity (herb, shrub, and tree layers combined) relative to the control.

However, differences in total structural diversity between treatments and control were,

for the most part, not significant (P > 0.10). Therefore, the vegetation management

treatments used in this study decreased only the volume of the targeted deciduous tree

layer and did not adversely affect the species richness, diversity, turnover, or structural

diversity of the plant community.

Table of Contents

Abstract ii

List of Tables vi

List of Figures vii

Acknowledgements viii

Introduction 1

Study Areas 6

Materials and Methods 7

Experimental Design 7

Vegetation Sampling 9 Crown Volume Index 11 Species Richness 11 Species Turnover 11 Species Diversity 12 Structural Diversity 12

Statistical Analysis 13

Results 15

Site Similarity 15

Crown Volume Index 15 Herb Layer 15 Shrub Layer 21 Deciduous Tree Layer 22 Coniferous Tree Layer 23

Species Richness 24

Species Turnover 26 Herb Layer 26 Shrub Layer 26 Tree Layer 28

Species Diversity 29 Herb Layer 29 Shrub Layer 29 Tree Layer 29

Structural Diversity 31 Herb Layer 31 Shrub Layer 31 Tree Layer 34 Total (herb, shrub, and tree layers combined) 34

iv

Discussion 35

Experimental Design 35 Crown Volume Index 36

Herb and Shrub Layers 36 Deciduous Tree Layer 38 Coniferous Tree Layer 39

Species Richness 42 Species Turnover 43

Herb Layer 43 Shrub Layer 44 Tree Layer 45

Species Diversity 45 Herb and Shrub Layers 46 Tree Layer 48

Structural Diversity 49 Herb Layer 49 Shrub Layer 50 Tree Layer 51 Total (herb, shrub, and tree layers combined) 51

Conclusions 52

Literature Cited 53

Appendix 1 63

Appendix 2 66

Appendix 3 68

List of Tables

Table 1. Repeated measures analysis of variance (RM-ANOVA) model used for

investigating plant volume, species diversity, and structural diversity 14

Table 2. Statistical results (RM-ANOVA) 16

Table 3. Crown volume indices (m /0.01 ha) of prominent herbs, shrub, and trees 17

VI

List of Figures

Figure 1. Design of strip-transects used to sample vegetation 10

Figure 2. Mean total crown volume index (m3/0.01ha) for herb, shrub, deciduous tree, and coniferous tree layers among control, manually, and cut-stump treated plantations 20

Figure 3. Mean species richness for herb, shrub, and tree layers among control, manually, and cut-stump treated plantations 25

Figure 4. Mean species turnover for herb, shrub, and tree layers layers within control, manually, and cut-stump treated plantations . 27

Figure 5. Mean species diversity indices (Simpson's and Shannon-Wiener) for herb, shrub and tree layers among control, manually, and cut-stump treated plantations .. 30

Figure 6. Mean Simpson's structural diversity indices for herb, shrub, tree, and combined total layers among control, manually, and cut-stump treated plantations 32

Figure 7. Mean Shannon-Wiener structural diversity indices for herb, shrub, tree, and combined total layers among control, manually, and cut-stump treated plantations. 33

vii

Acknowledgements

I thank the British Columbia Ministry of Forests, Research Branch, Victoria, B.C., and

the Salmon Arm Forest District for financial and logistical support. Additional funding

was provided by the Natural Sciences and Engineering Research Council of Canada. For

their assistance with field work, I also thank S. Allen, M. Burwash, B. Hammond, C.

Houwers, S. Milne, C. Nowotny, G. Ozawa, M. Porter, D. Ransome, E. Roberts, G.

Ryznar, and C. von Trebra. And finally, I thank L. Zabek, P. Marshall, and M . Pitt for

assistance with statistics.

viii

Introduction

Vegetation management is increasingly important in temperate forests of North America

as a means to increase the survival and growth rates of crop trees, and to more quickly

achieve free-to-grow status for regenerating stands (Newton and Comeau, 1990). If

allowable annual cut levels are to be maintained, accelerated development of new

plantations, as well as rehabilitation of backlog sites is required. Intensive silviculture

programs must deal with the problem of reducing competing herbs, shrubs, hardwoods,

and other non-commercial species to increase production of merchantable trees

(McDonald and Radosevitch, 1992). Vegetation management is, therefore, a tool which

can help provide rates of tree growth necessary to sustain the forest industry.

To accelerate the development of a second-growth stand of crop trees, vegetation

management is particularly important during the first few years following planting. This

is the time when a site is naturally dominated by pioneer species which, depending on the

objectives established for the site, may be considered as non-crop species. Species often

considered as non-crop competitors with coniferous crop trees are species of the genera

Betula (birch), Alnus (alder), Populus (poplars and aspens), Acer (maple), Prunus

(cherry), and Rubus (raspberry and thimbleberry). There are several methods for

reducing the competition created by these non-crop species such as 1) manual (hand-held

chain saws, brush saws, and girdling tools), 2) mechanical (machinery that mow, rake,

1

crush, and chip), 3) burning, 4) biological (use of grazing livestock and pathogens), and

5) herbicides.

Because the goal of vegetation management is to modify the plant community to favor

the rapid development of crop trees, it undoubtedly has profound effects on the

abundance of some plant species inhabiting treated areas (Santillo et al., 1989). Although

the benefits of vegetation management treatments on the growth and survival of

coniferous crop trees is well documented (Reukema, 1964; Crossley, 1976; Brix, 1981;

Berry, 1982; Harrington and Reukema, 1983; MacLean and Morgan, 1983; Maguire,

1983; Lavigne, 1988; Yang, 1991, Wang etal, 1995; Simard and Heineman, 1996a,

19966, and 1996c, and others), its effects on non-timber plant community attributes are

still not well understood (Wagner, 1993).

Because of growing public concern about the environment, forest managers are no longer

charged with only regenerating forests (Lautenschlager, 1986; McGee and Levy, 1988;

Freedman, 1991; Brand, 1992; Lautenschlager, 1993; Halpern and Spies, 1995). Today,

in addition to regenerating forests for future harvest, the conservation of biodiversity is

becoming an integral part of forest management. For example, the recent development of

the Forest Practices Code of British Columbia and the publication of a series of

guidebooks such as the Biodiversity Guidebook (B.C. Ministry of Forests, 1995), are

indications of this change within B.C.'s forest industry. However, as outlined by Wagner

(1993), there is a serious lack of information about the effects that vegetation

management has on the environment. Therefore, if forest management is to encompass

2

both timber production and the conservation of biodiversity, vegetation management

procedures must be critically examined within the context of both of these objectives.

The most common form of vegetation management within Canadian forests is aerial

spraying of the herbicide glyphosate (Vision®, commercial formulation containing

glyphosate, 356 g/L present as an isopropylamine salt) (Campbell, 1990). At prescribed

rates, glyphosate poses minimal toxicological risks, in terms of mortality and reduced

reproduction in wildlife, and does not accumulate in the environment (Morrison and

Meslow, 1983; Newton et al., 1984; Freedman 1991), however, public opinion about the

use of any forest herbicides is often very negative (Mitchell, 1990; Turpin, 1990;

Freedman, 1991; Glover, 1994). Research has shown that glyphosate does not have a

direct effect on the survival or reproduction of small mammals (Sullivan and Sullivan,

1981; Sullivan, 1990a; Sullivan et al., 1997). In addition, several studies have shown

that small mammal population responses to aerial application of this herbicide have

ranged from an overall increase in population density (Anthony and Morrison, 1985), to

no change in density (Sullivan and Sullivan, 1982; D'Anieri et al, 1987; Sullivan,

19906), to others which have reported a decrease in abundance (Clough, 1987; Santillo et

ah, 1989). Other studies have also reported that glyphosate does not have any significant

long-term effects on the survival or health of ungulates (Sullivan and Sullivan, 1979;

Cambell et al, 1981; Jones and Forbes, 1984). However, more information is still

needed to assess the effects that vegetation management treatments, particularly

alternative treatments (Campbell, 1990), have on both the timber resource and non-timber

values such as habitat quality.

3

One such alternative vegetation management treatment is the ground-based method of

applying glyphosate to the cut-stump surface of manually-cut deciduous trees. The non-

broadcast and species-specific approach of this treatment suggests that it may be an

effective and environmentally sensitive method of vegetation management. One study

reported that this cut-stump method, although significantly reducing crown volume index

of the targeted deciduous tree layer, had no apparent effect on the abundance of herbs,

shrubs, and coniferous trees, and did not affect the survival or reproduction of small

mammals (Runciman and Sullivan, 1996).

Studies designed to investigate the effects of vegetation management on biodiversity

often monitor changes in plant abundance and diversity (Tomkins and Grant, 1977;

Pollack ef al, 1990; Freedman et al, 1993; Sullivan, 1994; Sullivan et al, 1996; and

others). However, a habitat attribute that is frequently ignored by such studies is

structural diversity. Because of the well documented direct relationship between the

structural diversity of a habitat and the diversity of species that live there (MacArthur and

MacArthur, 1961; MacArthur, 1964; MacArthur, 1965; Balda, 1969; Sutton and Hudson,

1980; Adler, 1987; Harney and Dueser, 1987, Hunter, 1990; and others), structural

diversity should be carefully examined to determine the effects of vegetation

management treatments on habitat quality.

This study was designed to test the hypothesis that vegetation management treatments

(manual spacing and cut-stump applications of glyphosate) applied in young mixed-

4

conifer plantations would adversely affect the plant community (herbs, shrubs, and trees)

by decreasing 1) plant abundance, 2) species richness, 3) species turnover, 4) species

diversity, and 5) structural diversity.

5

Study Areas

This study was conducted within two similar areas in the Shuswap Highlands of south-

central British Columbia, Canada. The Eagle Bay (50° 55' N, 119° 11' W) and Sicamous

(50° 52' N , 118° 59' W) sites are both located within the Thompson Moist Warm variant

of the Interior Cedar-Hemlock biogeoclimatic zone (Lloyd et al., 1990). These areas are

characterized by similar topography, elevation, climate, and climax vegetation. The

topography of this area is hilly to steeply sloping and elevation ranges from 550 to 1237

m. Summers are generally warm and dry and winters cool and wet. Mean annual

temperatures range from 2 to 8.7° C and precipitation ranges from 500 to 1200 mm, with

as much as 50% falling as snow (Ketcheson et al., 1991). Climax forests on mesic sites

are characterized by western redcedar (Thuja plicata Donn ex D. Don) and western

hemlock (Tsuga heterophylla (Raf.) Sarg.), while interior Douglas-fir (Pseudotsuga

menziesii var. glauca (Beissin.) Franco), lodgepole pine (Pinus contorta Dougl. Ex Loud,

var. latifolia Engelm.), and paper birch (Betulapapyrifera Marsh.) are common serai

species. Nomenclature for all plant species follows Hitchcock and Cronquist (1973).

Nine plantations, ranging from 16 to 47 ha in area, and 2 to 9 years in age when selected,

were chosen in conjunction with the Salmon Arm Forest District silvicultural staff on the

basis of operational scale, proximity, and initial similarity in requiring vegetation

management treatments. The six plantations found at the Eagle Bay site were located on

a north to north-east facing slope. These areas were logged between 1978 and 1986 and

6

planted predominantly to lodgepole pine between 1985 and 1990. Site inspections

indicated that Douglas-fir, western larch (Larix occidentalis Nutt), and hybrid spruce

{Picea engelmannii Parry X P. glauca (Moench) Voss) were also present. The remaining

three plantations at the Sicamous site were located on a southeast slope. These areas

were logged between 1977 and 1988 and planted between 1982 and 1989, largely with

Douglas-fir. Inspections of these sites indicated that lodgepole pine was also present with

lesser amounts of hybrid spruce and western white pine (Pinus monticola Dougl.).

Materials and Methods

Experimental Design

This study follows a randomized-block design. Study plots were established in eight of

the nine plantations during September 1991, and in the ninth during April 1992.

Assignment of control and vegetation management treatments was done subjectively

among these nine plantations due to logistical and economic constraints. The first of

these constraints was related to the close proximity of the Eagle Bay plantations to a

residential community. To facilitate the approval of the herbicide application permits, the

cut-stump herbicide treatments had to be applied to the two plantations furthest from

these residents. The second constraint was related to a minimum acceptable outcome

anticipated for the cost of the vegetation management treatments. This required that the

more expensive treatment, the cut-stump herbicide treatment, be applied to the Sicamous

plantation that was most dominated with deciduous trees.

The nine plantations were stratified into three blocks on the basis of 1) location and 2)

elevation. Consequently, plantations A, B, and C (Sicamous site) formed Block 1, and

the remaining six plantations from the Eagle Bay site (D to I) were stratified into Blocks

2 and 3, on the basis of elevation.

In addition to untreated controls, the vegetation management treatments used in this study

were applied in the following manner. Manually treated plantations had all trembling

aspen (Populus tremuloides Michx.), paper birch, and black cottonwood (Populus

trichocarpa T. & G.) cut with power saws, except for a few individual stems that were

left in the openings of the plantation. A l l stems of willow (Salix spp.), Douglas maple

(Acer glabrum Ton-.), bitter cherry (Prunus emarginata (Dougl.) Walp.), and speckled

alder (Alnus incana (L.) Moench) that were within 1 m of a crop tree were also cut. A l l

other vegetation was left uncut.

Cut-stump treated plantations were treated in a similar manner as those receiving the

manual treatment, with the additional application of the herbicide glyphosate, diluted 2:1

in water, to the cut stump surfaces of the felled deciduous trees. A small amount of

Basacid Blue® dye was added to the herbicide mixture to mark the treated stumps. Most

maple, cherry, and alder stems were left uncut. These prescriptions were applied between

September 25 and October 22, 1992.

8

Vegetation Sampling

Vegetation was assessed annually within each plantation to determine the presence and

abundance of individual species for the purpose of monitoring the plant community's

composition and structure. Five permanent 5 x 25-m strip-transects, each consisting of

five contiguous 5 x 5-m subplots with nested 3 x 3-m and 1 x 1-m subplots (Figure 1),

were established within each of the nine plantations (as outlined in Stickney, 1980;

1985). Transects were randomly placed within each plantation as long as they were at

least 50 m from the nearest stand edge and they did not cross any roads, skid-trails, or

landings.

Within a strip-transect, each of the 3 different sized subplots were used to sample

2 2

different plant forms: the 5 x 5-m (25-m ) subplots for sampling trees, the 3 x 3-m (9-m )

subplots for sampling shrubs, and the 1 x 1-m (1-m ) subplots for sampling herbs (as

outlined in Ritchie and Sullivan, 1989). In addition, trees, shrubs, and herbs were each

sampled within 6 height classes of 0-0.25, 0.25-0.50, 0.50-1.0, 1.0-2.0, 2.0-3.0, and 3.0-

4.0 m (Walmsley et al, 1980). Species abundance was estimated within each of the 6

height classes that the plants occurred in by a visual estimate of percent cover. Individual

plants were only measured once within the height class containing the topmost growth of

that plant.

All plants can be classified into one of three life forms: herb, shrub, or tree. However,

these classes can be further subdivided or, conversely, grouped together, to create

additional classes of plants. For example, the tree layer may be subdivided into

deciduous and coniferous tree classes. Also, herbs, shrubs, and trees are sometimes

9

grouped together to form a class including all plant forms. Hereafter, the tree class w i l l

include both deciduous and coniferous trees and the total class w i l l include all plant

forms.

5 x 25 m Transect

Trees

Shrubs

Herbs

Figure 1. Design of strip-transects used to sample vegetation.

10

Crown Volume Index

A crown volume index was calculated for each species within each of the five nested

subplots of a transect by multiplying the percent cover values by the top of the

corresponding height class (Stickney, 1985). The product of these values gives the

volume of a cylindroid representing the space occupied by the plant in m3/0.01 ha.

Species Richness

Species richness (S) was calculated as the total number of species of a given plant form

(herb, shrub, or tree) sampled within a transect (Krebs, 1989).

Species Turnover

Species turnover (TO) was also calculated for herb, shrub, and tree layers. TO is defined

as the number of species lost and gained during a set time period, divided by the total

number of species sampled during the same period (Schoonmaker and McKee, 1988),

and is calculated as follows:

TO = / (A + B)

where: L is the number of species lost and G is the number of species gained during a

defined time period from ti to t2 and A and B are the total number of species sampled

during times ti and t2, respectively. Because a species turnover calculation requires a

11

minimum of two sample periods, this measure is undefined for the first year of sampling

(pre-treatment year).

Species Diversity

Two diversity indices are reported within this paper: Simpson's index, which is sensitive

to changes in abundant species (Simpson, 1949) and the Shannon-Wiener index, which is

sensitive to changes in rare species (Pielou, 1966a; Peet, 1974). Simpson's index of

species diversity is the probability of picking two organisms at random that are different

species, and ranges from 0 to almost 1. The Shannon-Wiener index of diversity is based

on information theory and the degree of difficulty in predicting correctly the next

individual sampled. As such, this index increases with number of species sampled, and

ranges from 0 to approximately 5 for biological communities.

Structural Diversity

The above definition of Simpson's and Shannon-Wiener diversity indices, although

referring to species diversity, applies equally to structural diversity. While species is the

object of a species diversity index, height class is the object of a structural diversity

index. The indices are calculated as follows:

Simpson's diversity index (D)

D = l-[Z(Plf]

Shannon-Wiener diversity index (H')

H' = -I(Pi)(log2Pi)

12

where: p, = proportion of mean total crown volume index (Pielou, 1966a) belonging to

the i'h species (for species diversity indices) or height class (for structural diversity

indices).

Pre-treatment vegetation sampling was initiated in late July 1992. The first post-

treatment sampling of vegetation was conducted in July 1993 and was conducted

annually to July 1996, when the study was terminated.

All plant community attributes (e.g., crown volume index, diversity indices) were

calculated on a subplot basis, and then averaged across the five subplots of a transect.

Consequently, each transect represents one datum.

Statistical Analysis

A repeated measures analysis of variance (RM-ANOVA, SPSS Institute Inc., 1997) was

used to test for significant differences among treatment means, as shown in Table 1.

Both pre- and four post-treatment years were analyzed together, resulting in five levels

(years) for the within-subjects factor (time). Both treatment and block were assigned as

between-subjects factors. Before performing any analyses, data not conforming to

properties of normality and equal variance were subjected to various transformations to

best approximate these assumptions required by any ANOVA (Zar, 1984). Mauchly's W

test statistic was used to test for sphericity (independence of data among repeated

measures) (Littel, 1989; Kuehl, 1994). For data found to be correlated among years, the

Huynh-Feldt correction was used to adjust the degrees of freedom of the within-subjects

13

F-ratio. The Bonferroni post-hoc test (adjusted for multiple contrasts) was used to locate

differences among treatment means within each sample year (Rosenthal and Rosnow,

1985). Significance levels for all analyses (i.e., RM-ANOVA and Bonferroni

significance tests) was set at a = 0.10. Although conclusions made from statistical

results with the more standard significance level of 0.05 may be stated with more

confidence than those resulting from tests with a probabilty of 0.10, significance tests

with P-0.10 are likely biologically significant and deserve comment. Tests regarding

data properties, such as Leven's homogeneity test and Mauchly's W sphericity test, were

made with a significance level of a = 0.05.

Table 1. Repeated measures analysis of variance (RM-ANOVA) model used for investigating crown volume, species richness, species turnover, species diversity, and structural diversity.

Source of variation

Factor type

Level Degrees of freedom F-test (d.f.)

Block Random n = 3 n-l=2 Treatment Fixed k = 3 k-l=2 MS-Treat. / MS-Err. I (2,4) Error I (n-l)(k-l)=4 Time Time X Treatment Error II

Fixed t = 5a t-l=4 (t-l)(k-l)=8 k(n-l)(t-l)=24

MS-T.xTreat. / MS-Err. II (8,24)

" There are five levels of time (t = 5) for all plant attributes except species turnover. Because species turnover is not defined for the pre-treatment year, there are only four levels (t = 4) for this attribute.

14

Results

Site Similarity

There were no statistical differences among the treatment and control plantations for any

of the plant attributes (crown volume index, species richness, species diversity indices,

and structural diversity indices) during the pre-treatment year (Table 2; Figures 2, 3 and 5

to 7).

Crown Volume Index

Herb Layer

Prominent herb species found within the study area included fireweed (Epilobium

angustifolium), white hawkweed (Hieracium albiflorum), common dandelion

{Taraxacum officinale), wild strawberry (Fragaria virginiana), and pearly everlasting

(Anaphalis margaritacea). The crown volume index of these and other prominent herbs

are provided, by treatment and sample year, in Table 3.

Although the mean total crown volume index (m /0.01 ha) of the herb layer changed over

time (Figure 2), differences among control and treatment plantations were not significant

(F2,4 = 1.26; P = 0.38). Both the control and treatment groups showed similar trends

during the four post-treatment years. The mean total crown volume index of the herb

layer for the control, manually, and cut-stump treated plantations peaked during the

15

Table 2. Statistical results obtained from repeated measures analysis of variance (RM-ANOVA) conducted on several plant community attributes. Five years of data (one pre-treatment year and four post-treatment years) were included in these analyses.

Attribute Treatment Effects Time X Treatment Interaction Attribute r(2.4> P P

Volume Index Herbs 1.26 0.38 F(6,19) = 0.73 0.63

Shrubs 0.86 0.49 F (8,24) = 1.15 0.37 Deciduous trees 40.63 0.002 F(6,19) = 9.94 < 0.001 Coniferous trees 0.99 0.45 F(8,24) = 0.33 0.94

Species Richness Herbs 3.29 0.14 F(8,24) = 1.77 0.13

Shrubs 2.75 0.18 F(8,24) = 0.31 0.95 Trees (decid. and conif.) 1.17 0.40 F(8.24) = 0.85 0.57

Species Turnover Herbs 0.72 0.54 F(6,18) = 0.37 0.89

Shrubs 1.66 0.30 F(6,18) = 1.24 0.33 Trees (decid. and conif.) 2.14 0.23 Fr6.i8i= 1.06 0.42

Species Diversity - Simpson's Herbs 3.34 0.14 F a ,22)= 0.34 0.93

Shrubs 1.26 0.38 F(7,22) = 0.29 0.95 Trees (decid. and conif.) 1.49 0.33 F(7.22) = 2.66 0.04

Species Diversitv - Shannon's Herbs 2.78 0.18 F(7,22) = 0.24 0.97

Shrubs 1.17 0.40 F(7,21) = 0.36 0.92 Trees (decid. and conif.) 1.22 0.39 F(7.23) = 2.70 0.03

Structural Diversitv - Simpson's Herbs 0.09 0.92 F(8,24) = 0.24 0.97

Shrubs 0.05 0.95 F(8,24) = 0.62 0.75 Trees (decid. and conif.) 3.68 0.12 F(7,23) = 1.73 0.15

Total (herbs, shrubs, and trees) 6.34 0.06 F(7,23) = 2.39 0.05

Structural Diversitv - Shannon's Herbs 0.16 0.85 F(8,24) = 0.30 0.96

Shrubs 0.09 0.92 F(8,24) = 0.70 0.69 Trees (decid. and conif.) 3.70 0.12 F(7,23) = 1.73 0.15

Total (herbs, shrubs, and trees) 7.09 0.05 F(7.22) = 1.60 0.19

a Degrees of freedom for tests of Time x Treatment interactions are 8 and 24 (or 6 and 18 for species turnover) when the data are not correlated among years. Correlation among repeated measures data is a violation of ANOVA assumptions and requires an adjustment (Huynh-Feldt correction) which decreases the df from those mentioned above. Note: Significant P-values (a = 0.10) are indicated in bold text.

16

3 2 s U

ro

2 -

CO

m

xf xf IO X -

o p

CO x f i n T~

o P

00 o o o ) x-i n O O ; - ^ i n

01 x~ c\i p

cn CM CD N. CO p

CM CO

in co

T- CNJ

CO CO co x f

T- P

CM K -CM O O p

T - CO XT x-O p

CM N~ xr CM

d P

CM CO CM T~

o p

CM in CO -<--

o p

CO T - O

o p

•r- Xf

o p

CO CM m CM o p

CM co cn to P

CO

CO

00

CO i n co

pi CM

xr co

t- 00

XT p

CO CO xf o o p

00 CO P

o p

CD CO CN <M T - C j

xr m h- x~ d P

co oo CD x— o p

oo xj-T - O

o p

CD CO

o p

CD CO CM O

o p

x f 00 CM O

o p

CO f~ o p

h - 00

o p

xf CO

ci p m x-o o o p

o d

in" c n 1

o CM o o O o

Xf i n CMCM 00 O O O p

CO

CD x— CM x-O P

co in T - O

d p

CO CM o o o p

xf oo CM o o p

CD CM o o o p

o o P

00 l O : N 'SI-CO x- |o P

o p o p oo

i o P i d p

cn oo CO X -

o p

co oo co x-d p

in xf x-d p

CO X -CO CM

d p

CD x-CM x-d p

CD 00 O P

d p

CO CM o o d p

T- 00 CM P

d p

d p

cn xt-o o d p

CM Xf T - p d p

co oo CM x-d p

CO CM

T— X-d p

m xf o P d p

cn o o i ^ t in o 0|T- P d p d p

co co

d p

oo

d p

CO IS-T - O

d p

CD oo CM x-d p

i n CM o o d p

xr oo

d p

m CM o P

d p

CM o o d p

co

d p

in CM o o d p

oo oo o o d p

o o d

CM CM O O

d p

i n oo o o d p

co in o o d p

oo o o d p

CO i i n CM; d p;

co CO T~

d p

cn o> CM t~ d p

co

CD CM

CM cn

^ P

00 00

in in CM

CM x~ in

d p

co CM

d p

oo o in CO —•

CM xr CO

CO*

o CM iri oo d p_

CO*

in in1

o CO CO

CO d

CO

CM

co co p CO J ~

co m cn P

CM

co oo

CO* 00 CO CM*

CM oo d O

CM* 00 in o

CO* 00

o

CM CM

CO XJ-Xf d p

oo oo CM O

d p

m oo CM o d p

CM Xt-

d p

00 f > C M o> T - | ^ i o o oo

d p

CO N -cn o CM

co oo

CM p

CM <o CM N-CM P

d p

0 1 CO* co ^ " o o -<~-CO XJ-ICO t~ T - : & o o ^ ^ - c o !c i o p

co x}-cn co

m

CM in;CM co CM' ^ i t M P

T- CM o P

S2 CM* co 5) CM 52.°' S.

00 CM

cn oo oo cn

P

oo

•3- m ^ 00

in CM

O

U

CO Oi CM 00

CD P

m oo CD 00

XT P

00 N . CM CM

in co

CM N -

xr P

in ^ m in CM p

CD CO

d p

in x~ d p

oo XJ-co oo

T- p

cn co m x~ d p

ani

.a

cie

on Spe

cu -*-»

Spe

OH CO o

s -Q O

5

T - XJ-: T - P i

d p

co CM O

d p

m in CM

d p

xf CO

o co d p

Xf p

CO x~ o o d p

m x f o o d p

CO i n x~

m xf xr

d p d p

co CO x-d p T -

T— 1 00 CM

d p

CO

p

CO 00

T- P

d p

oo co oo oo

d p

d p

CM x~ d p

.CO TO

CO

c

CO 00 CM 00

CM oo CD x~ d p

00 CO xf x~ d p

•<- co CM

d p

xf CM in CM

d p

CD CM o o d p

00 oo o o d p

CD in d p

co m

d p

CM CO

.co

£ g

s §

co in CO x~ d p

CO*

o CD x?

00 CM CO*

o d o d

^* p CM CO

cn* o m o CO*

o d o d

C 3

CO

CD £

i - P d p

^_ Xf

° P

co cn co o d p

•r- CO CD Xf xf j - ;

T - CO cn oo

CM p

co oo cn CM p

CM m CM oo

co p

cn oo o in CM p

w

ro

S8M3H

S .-2 ^ ci co

co; 3 >Ci-Q

."3 ><

CO* CO 00* f ^ i C M CO K T- Ci

c^iSS io p

^ xf ,m i -°° inicn m

CM 5SJ°' S

. CM;CM CM 1 - Xf ;cO in

OOi

oo oo xf cn 00 X f

CO 00 h - CO

xf CM

I— CD 0 0

00

00 CM o 00

CM P

i n N-i n o

to 5; j g : CO CO ^ i 3 ? £ i - Q CD r~ ~^ m CO ^"3 co i"3 i2 2 P

0. EjQ: Q i Q ; B L t IB SflflHHS

CO

c 3

17

£> o coco \ 0\ • • C i — CO i d

a. E

r o co 5J1 5? R gj |00 COCO LO •<- o

d C M T - 1 p O p

a

0\

OS

O * H

U r o

CD co co <o ~ CSI T ~ o

o p o p

CO CO X t - C O xj- Oi CN I ^ T - ; x f i o O v - O T - 1 p ; - * - 1 p o p ' ; 0 P

i - co CN xj* CD co" in oJ

d p

co cn o p o p

O IO T - 0> O o CO P O P

CO

d p

x~;o d P : d

CN •" tcoT -cd^mio co co co i i - o in co co j ^ d p d p ' i d c i

C M

d

o i : co i^- io CN

CM co x- P

d p d p

CNCD x ~ o o M-

O x~ TJ- C N i O o co x - - i d p ' i d p

C M m CM IO CM P

h- Oi CO x-

d p

CT) oo;co o p d

CN CM CD X } -T f CO CM x-d p d p

CM CO CM 1

CM o O

P_ d

X f o 3"

p ' d — d „ :

CM 00 ioi X ~ o o

o_ d

x~ o !

x- CN CN K. CD •«-•

T f co o x-

co x~ TT CN d p

co M -x r N .

cn co CN ^ co

r - io CD x-

o p

g Ri^r In-xr C Q C D ifj

• 5° ° p CN *~. T f CO

CN

O p

T f X J -• CM

CO

^ S ^ T - , ~ in CN <=>. v. r-~ <°.

CO CN O P

o p

m x- T f oo CO COjCO T -

d p ' i d p

in

° ? LO C O

CO M- CO O O T - P

d p d p

•CM CJ)

CM co co x-d p

r - o> CN o d p

o R p CO CO

CM" CN CO

CM d P_ d o d

CD O IO m CD

UO CD

CO d d o d

t- CO CO co CN lOi^f r-i

p d p ' i

CN

p

co in Oi m co p ' i d p

i co in i co co

• )V i^f t~ CN f~

co l O i c o co CN ^ :

* : p i d S .

co o o d p

m co o P

d p

m CM o o

o P

o P o P

CO CO co

d p

oo

CN -<-;co c o : O P CO CNi o p

T - CM t~

d p ' i d p

o p

CO; co t o i d p i

CD oo in r-CD

^ CN

3 ! < 5^

C O co CD \ h CD

C0 in a T™• to CO C

CN; .

p'i£ !

CO

i d i 1 ^

S co~i io" CM 1 0 "!HCD •y-iCD CN

cb cd co d p

CM CD co CN O 00

00 CM P

m co CM co in ^ ; c o d p oo co

oV; in 5 s T - co" S co~i co in |^ C0;^f 'TfiCM , r> ^ L O

cd co i cd d p

CO CO io io T-^ p

CD CM . . °H OiCD 1

2 od in CM

^ co" "f- T _

io oo g o c ^ £ ^ = ^ ?

P2 coico . O i O CM

i ° P CD CM

2 S N CM

g CM

S F0 T - p

CO ^ ;

io~ oo oTi CD ?J . CO CO ,

II co" CD co d

t— co N C \ | r - O CM J~ j -"

IOOOSO) oi m T - p d p

r~- co •^•in N-co CD CM

h~ o o i m c*51io 6T CD c o i m o i c o t o

d p CN v-^id p

O) CD CSliT- oo i H - S M I O co co T -roi-<3-co

to T t CO CM <M

!CD CNitn CN

CO CN 00

CN T f

iri IO iri

CD 00

T f OO

m co

CD T f

5? 00

CO CO

CO m

o 1 CM

CO T ? CO co

T f T f

TT CN

w o o CM

WJ CN l ^ C N

N - i o com

o o o p

co co l O i ^ in T - : p

Pio

uo i n 1 0

. C O CO

. CO

° p

IO

C M

•*J; I^ CN : CD CO uu T J - i r - T f : ! ^ . LO OR LO

00 N . ; CD i o :

T - CO CO

K CN p

00

00 x f r - l T f C\j

cd~ Oi

oo l ed

0>;CO Xh! COjT- x~

CD CO

CD

CO CNicO

co ICN

c o 1

Oi oo CO

R Oi

T ~ o

co1 T—

o_ T— p ^

00 CD

o CO"

IO

CD CO T—

T f J -

<3' oo co* 1 0 P 0 0 ° ^ co °^ Q xt-OO ^ ; 0 0 N . C D O)! ° ? CO

co 2i i°2 5 2 J v ^ p co ^ LO

CO CD CN

co

• P: • P

CD 00 CO X f O xf T f CO

x-" p i d p

u

CD O - .S O

CO .CO

"3 5 !fl w

£ w *

« a: cB oc -a <c

^•8 Q; .<Q DC .CD

(panunuoD) saHMHS

CD £ « '55

.to

•2 CO 3 O

- Q O 3 3

5 ct

CD

3 o o y Q, S

CD -2 CD

to o

s;i=

to

P CD

-2

•c; o, i" co

,to o

co

3 CO

X

CD

S33HI

18

5 . 5

a

a E = 2

= 5 « 2

£ x* e g ! o — U

u C/5

L O XT CN

00 N -m x -

CN co 00 CN

R* C N C O " C N

cn* C O o o o C N T— X—• d ci_ o j d

I O C O

o o d p'

o d

C D X f d p

C O * m CO* xf •<— o xf Xf CN Csi o d oj d oj

C O * — •xf 5>* x f C O C O o o C N X— T— V -

pj d oj d oj d pj

>o oo*i S 3* cn co* co co; g § ^ CSj

csi j ~ | _; p d p

in xf T—

d p

C N

o

xf p CN CO o O p j d O j

cn K. o O d p

CN co* CN o 00 o P pj in coj d Pj

CN x -O O d p

o o d p

C N C O

C O 9 x - C O

CN x~ o o d p

o o d p

o o d p

S* o C D

N -C O * C N o o

oj x- d

C N in o

C O

oo o C N

o O C O x-- d

C O * O C N

co* O o o

Oj T~ x- d

55* C O

C O

oo oo* C O o o

x- d c j d

r- xf o P d p

x - in o N. T - P

CD O

< o £ •S o Q: S

.CO 1 c

.x S C O CO o

- J o

o p

CO 3 s iS CD

,co cf (panuiiuoD) saajx

c/3 c o

fl"

o o o CD

, f l

fl

3 T3

19

Control Manual Cut-stump

Mean Total Crown Volume Index

Herbs Shrubs

40 <?30 1 20

'92 | '93 '94 Yea

'95 '96 '92 '93 '94 Yea

'95 '96

Figure 2. Mean total crown volume index (m /O.Olha) for herb, shrub, deciduous tree, and coniferous tree layers among control, manually, and cut-stump treated plantations. No statistical differences (a = 0.10) were observed for herb, shrub, or coniferous tree volume, however, the deciduous tree layer was significantly affected by the treatments. Arrow on horizontal axis indicates timing of treatments. * P < 0.10, **P< 0.05, *** P < 0.01; significance by Bonferroni post-hoc test.

20

second post-treatment year (1994) and then gradually decreased during the final two

years of the study. Both the manually and cut-stump treated plantations showed a

decrease in volume of herbs in the first post-treatment year, in contrast to the increase in

herb volume observed within the control during this same period, although differences

were not significant (Figure 2).

Shrub Layer

Prominent shrub species found within the study area include falsebox (Pachistima

myrsinites), thimbleberry (Rubus parviflorus), red raspberry (Rubus idaeus), bitter cherry

(Primus emarginatd), and willow (Salix spp.). The crown volume index of these and

other prominent shrubs are provided, by treatment and sample year, in Table 3.

Although the mean total crown volume index of the shrub layer changed over time

(Figure 2), differences among control and treatment plantations were not significant (F2;4

= 0.86; P = 0.49). The control plantations showed a steady increase in mean total crown

volume index of shrubs throughout the five years of the study. Both the manually and

cut-stump treated plantations also increased in shrub volume during the post-treatment

years. Like the herb layer, the shrub layer experienced a slight decrease in volume during

the first post-treatment year within the treated plantations; however, differences were not

significant (Figure 2).

21

Deciduous Tree Layer

The three deciduous tree species that were found within the study area were paper birch,

black cottonwood, and trembling aspen. The crown volume index of these deciduous tree

species are provided, by treatment and sample year, in Table 3.

The mean total crown volume indices of deciduous trees were not statistically different

(Bonferroni; P = > 0.17) among the control and treatment plantations during the pre-

treatment year. The treatments resulted in siginificant (F2,4 = 40.63; P < 0.002) decreases

in deciduous tree volume during the first post-treatment year (Table 2; Figure 2). In the

first post-treatment year, the mean total crown volume index of deciduous trees decreased

by 87% (106.5 to 13.5 m3/0.01ha) and 95% (46.8 to 2.2 m3/0.01ha) within the manually

and cut-stump treated plantations, respectively. This decline was in direct contrast to the

control plantations, which experienced an increase in mean total crown volume index of

38% (53.2 to 73.3 rrrVO.Olha) during this same period (Figure 2). Consequently, both the

manually and cut-stump treated plantations had significantly less (Bonferroni; P = 0.05

and P < 0.01, respectively) deciduous tree volume than that of the control during the first

post-treatment year. In addition, during the first post-treatment year, the cut-stump

treated plantations had significantly less (Bonferroni; P = 0.02) deciduous tree crown

volume than those treated manually (Figure 2).

Both the control and cut-stump treated plantations showed a gradual increase in mean

total crown volume index of deciduous trees throughout the four post-treatment years.

The dramatic suppression of deciduous tree volume observed one year after treatment is

22

maintained during the post-treatment years for the cut-stump treatment. As a result,

deciduous tree volume within the cut-stump treatment is less (Bonferroni; P < 0.02) than

that of the control during all four post-treatment years. The suppression of deciduous tree

volume within the manually treated plantations was, however, only short-lived. There

was a rapid increase in growth by the second post-treatment year (1994), resulting in

deciduous crown volumes becoming similar (Bonferroni; P = 1.00) to that of the control

during the 1994 sample year, and thereafter (Figure 2).

Coniferous Tree Layer

Prominent coniferous tree species sampled within the study area included interior

Douglas-fir, western redcedar, lodgepole pine, western hemlock, hybrid spruce, and

western white pine. Western larch, subalpine fir (Abies lasiocarpa), and western yew

(Taxus brevifolia) were also found within the study area, but were much less common

than the coniferous tree species listed above. The crown volume index of these

coniferous tree species are listed, by treatment and sample year, in Table 3.

Although the mean total crown volume index of the coniferous tree layer changed over

time (Table 3, Figure 2), differences among control and treatment groups were not

significant (F2,4 = 0.99; P = 0.45). The control plantations showed only a slight increase

in mean total crown volume index of coniferous trees throughout the five years of this

study. However, both the manually and cut-stump treated plantations exhibited

accelerated growth rates of coniferous tree volume relative to that of the control during

23

the post-treatment years. During the four post-treatment years, the mean annual

percentage increment of coniferous crown volume (see formula below) was 5%, 44%,

and 29% for control, manually, and cut-stump treated plantations, respectively.

where: A is the crown volume at time 1, B is the crown volume at time 2, and C is the

number of years between time 1 and 2.

Although the differences in coniferous tree crown volume among treatment and control

plantations were not statistically significant, increased coniferous growth rates of 5 to 8

times that of control plantations suggested a biologically and likely economically

significant treatment effect.

Species Richness

During this study, a total of 75 herb, 37 shrub, and 12 tree species were sampled. A list

of these herb, shrub, and tree species is provided in Appendices 1, 2, and 3, respectively.

There was no statistical difference in mean species richness among the control and

treatment plantations for herbs (F2,4 = 3.29; P = 0.14), shrubs (F2 > 4 = 2.75; P = 0.18), or

trees (F2,4 = 1.17; P = 0.40), at any time during this study (Table 2; Figure 3). Some time

trends, common to both control and treatment groups did occur. Although not

Mean annual percentage increment of crown volume

24

Control Manual C U Cut-stump

Species Richness

Herbs Shrubs

I '94 Year

'95 '96

Trees

Figure 3. Mean species richness for herb, shrub, and tree layers among control, manually, and cut-stump treated plantations. No statistical differences (a = 0.10) in species richness were observed among the control and treatment plantations for any of the plant forms (herbs, shrubs, or trees) during any of the five sample years. Arrow on the horizontal axis indicates timing of treatments.

25

statistically significant, the observed peak in mean herb species richness two years post-

treatment (1994), was followed by a gradual decline until the completion of the study

(Figure 3). The mean species richness of the shrub layer, although more static than that

of the herbs, also peaked in the second and third post-treatment years (1994 and 1995)

(Figure 3). The mean species richness of the tree layer was the most static of all, with

only very minor differences observed from the start to the end of this 5-year study

(Figure 3).

Species Turnover

Herb Layer

There were no significant (F2,4 = 0.72; P = 0.54) differences in herb species turnover

observed among control and treatment plantations during any consecutive year intervals

(Table 2; Figure 4). The lowest species turnover was observed between the pre-treatment

and first post-treatment year (1992 to 1993), and the highest was observed the next year

(1993 to 1994). Control, manual, and cut-stump treatments all showed similar time

trends throughout the study.

Shrub Layer

There were no significant (F2,4 = 1.66; P = 0.30) differences in shrub species turnover

observed among control and treatment plantations during any consecutive year intervals

(Table 2; Figure 4). Although not statistically significant, differences in shrub species

turnover among treatment and control plantations do suggest some trends. Both manual

26

Control I I Manual CZ] Cut-stump

Species Turnover

Herbs

Shrubs

0.1 0.08

o0.06 H 0.04

0.02 0

Trees

'92-'93

t

'93-'94 '94-'95 Time interval (years)

'95-'96

Figure 4. Mean species turnover for herb, shrub, and tree layers within control, manually, and cut-stump treated plantations. No statistical differences (a = 0.10) in species turnover were observed among the control and treatment plantations for any of the plant forms (herbs, shrubs, or trees) during any of the five sample years. Arrow indicates timing of treatment.

27

and cut-stump treated plantations experienced the highest species turnover during the pre-

treatment and first post-treatment time interval (1992 to 1993) and then declined

thereafter. This is in contrast to the shrub species turnover observed to be its lowest

within the control plantations during this same time period, followed by an increase

during consecutive years, peaking during the 1994 to 1995 time interval.

Tree Layer

There were no significant (F2,4 = 2.14; P = 0.23) differences in tree species turnover

observed among control and treatment plantations during any consecutive year intervals

(Table 2; Figure 4). The small changes in tree species turnover observed during the

course of this study were caused by a single species being gained or lost within a given

time interval. An exception to this was observed during the pre-treatment to first post-

treatment year interval (1992 to 1993) when, within one of the three manually treated

plantations, two tree species were gained.

By the final post-treatment year (1996), no tree species present during the pre-treatment

year had been lost within any of the control or treatment plantations. However, during

this same time interval, some species were gained. Western yew, western white pine, and

western larch were gained within the control, manually, and cut-stump treated

plantations, respectively (Table 3).

28

Species Diversity

Herb Layer

At no time during this study did either the Simpson's (F^ = 3.34; P = 0.14) or Shannon-

Wiener (F2,4 — 2.78; P = 0.18) mean species diversity of the herb layer differ statistically

among the control and treatment plantations (Table 2; Figure 5). Both Simpson's and the

Shannon-Wiener index of diversity showed similar time trends among the control and

treatment plantations. Specifically, the species diversity of the herb layer appeared to

peak one year post-treatment (1993) and then gradually decline during the remaining

three years of the study (Figure 5).

Shrub Layer

At no time during this study did either the Simpson's (F2>4 = 1.26; P = 0.38) or Shannon-

Wiener (F2 j4 = 1.17; P = 0.40) mean species diversity of the shrub layer differ statistically

among the control and treatment plantations (Table 2; Figure 5). Both control and

treatment plantations demonstrated the highest level of shrub species diversity during the

pre-treatment year and showed a gradual decline in diversity thereafter.

Tree Layer

At no time during this study did either the Simpson's (F^ = 1.49; P = 0.33) or Shannon-

Wiener (F2 ;4 = 1.22; P = 0.39) mean species diversity of the tree layer differ statistically

among the control and treatment plantations (Table 2; Figure 5). The greatest change in

both Simpson's and Shannon-Wiener tree species diversity indices occurred between the

pre-treatment and first post-treatment year and accounts for the significant (¥7,22 = 2.66; P

29

Control Manual E 3 Cut-stump

Simpson's Index

Herbs

Trees

Species Diversity

2 1.5 1

0.5 0

Shannon-Wiener Index

Herbs

'92 '93 '94 Year

'95 '96

Trees

'92 | '93 '94 Year

'95 '96

Figure 5. Mean species diversity indices (Simpson's and Shannon-Wiener) for herb, shrub and tree layers among control, manually, and cut-stump treated plantations. No statistical differences (a = 0.10) in species diversity were observed among the control and treatment plantations for any of the plant forms (herbs, shrubs, or trees) during any of the five sample years. Arrow on horizontal axis indicates timing of treatments.

30

= 0.04 and F7 23 = 2.70; P = 0.03, respectively) Time x Treatment interactions reported in

Table 2.

Structural Diversity

Herb Layer

The mean structural diversity of the herb layer for both the Simpson's (F 2 4 = 0.09; P =

0.92) and Shannon-Wiener (F2,4 = 0.16; P = 0.85) indices were statistically similar

among the control and treatment plantations throughout the study (Table 2; Figures 6 and

7). The control plantations were observed to peak in herb structural diversity in the first

post-treatment year (1993) and gradually decreased thereafter. Similarly, the plantations

that received treatments were also observed to peak in structural diversity, however, this

peak lagged behind that of the control by one year (i.e., peaked during the second post-

treatment year).

Shrub Layer

The mean structural diversity of the shrub layer for both the Simpson's (F2,4 = 0.05; P =

0.95) and Shannon-Wiener (F2,4 = 0.09; P = 0.92) indices were similar among the control

and treatment plantations throughout the study (Table 2; Figures 6 and 7). The structural

diversity indices of the shrub layer changed very little during the five year course of this

study. The only notable exception was a slight decrease in structural

31

Control [ZD Manual Cut-stump

Simpson's Structural Diversity

Herbs Shrubs

Trees Total

Figure 6. Mean Simpson's structural diversity indices for herb, shrub, tree, and combined total layers among control, manually, and cut-stump treated plantations. Bonferroni post-hoc tests indicated no statistical differences (a = 0.10) in structural diversity among any of the control and treatment plantations within the herb and shrub layers, pre- and post-treatment. However, some statistical differences among the treatment and control groups were observed within the combined total layer. Arrow on horizontal axis indicates timing of treatments. * P < 0.10, ** P < 0.05, *** P < 0.01; significance by Bonferroni post-hoc test.

32

Control Manual I I Cut-stump • •

Shannon-Wiener Structural Diversitv

Figure 7. Mean Shannon-Wiener structural diversity indices for herb, shrub, tree, and combined total layers among control, manually, and cut-stump treated plantations. A Bonferroni post-hoc test suggested no statistical differences (a = 0.10) in structural diversity among any of the control and treatment units within the herb, shrub, and tree layers, pre- and post-treatment. However, some statistical differences among the treatment and control groups were observed within the combined total layer. Arrow on horizontal axis indicates timing of treatments. * P < 0.10, ** P < 0.05, *** P < 0.01; significance by Bonferroni post-hoc test.

33

diversity observed immediately following treatment within both manually and cut-stump

treated plantations.

Tree Layer

The mean structural diversity of the tree layer for both the Simpson's (F^ = 3.68; P =

0.12) and Shannon-Wiener (F2,4 = 3.70; P = 0.12) indices were similar among the control

and treatment plantations throughout the study (Table 2; Figures 6 and 7). The structural

diversity of the tree layer decreased for both the control and cut-stump treated plantations

from the pre-treatment year through to the final year of sampling. The manually treated

plantations increased slightly in tree layer structural diversity in the first post-treatment

year, followed by a gradual decline during the final three years of sampling. The rate of

decrease in tree layer structural diversity (both Simpson's and Shannon-Wiener) was

greatest for the control and least for the cut-stump treated plantations.

Total (herb, shrub, and tree layers combined)

The mean total structural diversity indices (both Simpson's and Shannon-Wiener) for the

control and treatment plantations gradually declined following a single peak. However,

the year in which this peak occurs differs among control, manual, and cut-stump treated

plantations (1992, 1993, and 1994, respectively) (Figures 6 and 7). Differences in total

structural diversity among control and treatment plantations were indicated by a

significant (F2,4 = 7.09; P = 0.05) treatment effect for total Shannon-Wiener structural

diversity (Table 2). During the second post-treatment year, total Shannon-Wiener

structural diversity within both manual (Bonferroni; P = 0.08) and cut-stump treated

34

plantations (Bonferroni; P = 0.01) was greater than that of the control plantations (Figure

7).

A significant Time X Treatment interaction for mean total Simpson's structural diversity

(F723 = 2.39; P = 0.05) made it difficult to comment on the significant main treatment

effect. However, a significant treatment effect (F^ = 7.37; P = 0.04) without interaction

(F6,i8 = 1.63; P = 0.20) was observed when a RM-ANOVA was performed on the post-

treatment years only. This suggested that the interaction was caused by a change in total

Simpson's structural diversity among control and treatment plantations from the pre- to

post-treatment year and that a treatment effect did exist during the post-treatment years.

Post-hoc tests indicated that, as with the Shannon-Wiener index, total Simpson's

structural diversity was significantly greater than that of the control within both the

manual (Bonferroni; P = 0.02) and cut-stump treated plantations (Bonferroni; P < 0.01),

during the second post-treatment year (Figure 6).

Discussion

Experimental Design

The subjective allocation of the cut-stump herbicide treatment was a potential concern

because this was not consistent with a randomized-block design. However, although

natural variation was apparent among the plantations, the statistical similarity of control

and treatment units during the pre-treatment year (Figures 2 to 7) suggested that the

35

subjective allocation of cut-stump treatments did not significantly bias the experimental

design of this study. In addition, treatment allocation was well interspersed among the

nine plantations (and among the three blocks), a condition that is considered by some to

be more important than randomization. Hurlbert (1984) states that ".. .interspersion is the

more critical concept or feature; randomization is simply a way of achieving interspersion

in a way that eliminates the possibility of bias and allows accurate specification of the

probability of a type I error". This is especially true when replication is low, as is often

the case with large-scale ecological studies.

Crown Volume Index

Herb and Shrub Layers

Because the manual and cut-stump treatments targeted only specific deciduous tree

species, it is not surprising that there was not a significant effect on the crown volume

index of herbs or shrubs (Table 2; Figure 2). However, these treatments appeared to

temporarily depress the growth of these plant forms, relative to the control. Within the

control plantations, during the first post-treatment year (1993), the mean crown volume

index of the herb layer increased by 59% from the pre-treatment year (12.3 to 19.6

m /O.Olha). Whereas, during the same period, decreases of 24% (10.8 to 8.2 m /O.Olha)

and 17% (13.2 to 11.0 m70.01ha) were observed for the manual and cut-stump

treatments, respectively. Similarly, from 1992 to 1993, the mean volume index of the

shrub layer increased by 11% (26.6 to 29.6 m /O.Olha) within the control plantations.

The shrub layer decreased in volume by 50% (36.7 to 18.5 m3/0.01ha) and 19% (28.5 to

36

23.1 m3/0.01ha) within the manual and cut-stump treatments, respectively, during this

period.

The small and temporary decrease in abundance of understory vegetation (both herbs and

shrubs) noted in response to treatments during this study was in contrast to the significant

decreases in understory biomass reported in several studies treated with broadcast

applications of glyphosate (Pollack et al., 1990; Sullivan, 19906; MacKinnon and

Freedman, 1993; Sullivan, 1994; Simard and Heineman, 1996a; 19966; 1996c; Sullivan

et al, 1998; Whitehead and Harper, 1998; and others). This suggested that, relative to

the more commonly prescribed broadcast applications of herbicides, the species-specific

approach of cut-stump herbicide applications appeared to have minimal effects on the

understory vegetation biomass.

From the second post-treatment year (1994) to the final year of sampling (1996), both

herb and shrub crown volume indices fluctuated in a similar manner among the control

and treatment plantations presumably in response to natural successional processes.

The deciduous tree slash resulting from the vegetation management treatments likely

resulted in the short-term decrease in herb and shrub volume observed one year after

treatment. The physical obstruction caused by the deciduous tree slash may have 1)

impeded the growth of both herbs and shrubs, and 2) damaged some perennial shrubs.

The expected recovery of both herb and shrub volume observed by the second post-

37

treatment year (1994) was likely due to plants taking advantage of the increased

resources (light and moisture) created by the treatments.

Deciduous Tree Layer

Because treatments targeted the deciduous tree layer, the dramatic decrease in mean

deciduous crown volume was expected in the first post-treatment year (Figure 2). Both

the manually and cut-stump treated plantations had significantly less (Bonferroni; P =

0.05 and P < 0.01, respectively) deciduous tree volume than that of the control during this

first post-treatment year. However, rapid regrowth of paper birch and trembling aspen

trees via prolific stump sprouts occurred within manually treated plantations.

Consequently, by the second post-treatment year (1994), deciduous tree volume within

the manually treated plantations had returned to levels similar to that of the control.

Other trials have also recorded the vigorous sprouting ability of paper birch and other

hardwood species following manual cutting (Hart and Comeau, 1992; Simard and

Heineman, 1996a). As reported in other studies (Johansson, 1985; Marrs, 1985), the

treatment of cut-stumps with the systemic herbicide glyphosate limited the sprouting

behavior of hardwoods, resulting in continued suppression of the deciduous tree layer for

the four post-treatment years of this study. Relative to the volume of deciduous trees

within the control plantations, the manually treated plantations had a treatment effect

which lasted only one year.

38

My findings and those of other studies (Christensen, 1984; Lund-Heie, 1984; Johansson,

1985; Marrs, 1985; Wall, 1990), suggest that treatments with the objective of suppressing

deciduous tree species capable of growing prolific stump sprouts, such as paper birch and

trembling aspen, should employ measures to impede this sprouting behavior. My results

indicate that the application of glyphosate to cut-stumps of paper birch and trembling

aspen is an effective approach to reduce competition of these species with coniferous

crop trees. In addition, the species-specific nature of the cut-stump method appeared to

benefit the growth of coniferous crop trees and had no significant effects on other plant

community attributes, such as species richness, species turnover, species diversity, and

structural diversity.

Coniferous Tree Layer

Although the sampling methods employed during this study did not measure specific

attributes of coniferous tree growth (such as diameter at breast height and exact heights),

the mean crown volume index did measure the growth response of the entire coniferous

tree (branches and bole). Therefore, the crown volume index of the coniferous tree layer

is likely directly related to coniferous bole production, which is of primary importance to

the silviculturalist gauging the success of vegetation management treatments.

Although no statistical differences in coniferous crown volumes were observed among

control and treatment plantations, pre or post-treatment, trends in growth response during

the post-treatment years do suggest a treatment effect (Figure 2). The mean annual

39

percentage increment of coniferous tree volume (accumulation of volume) within the

control units was very gradual during the four post-treatment years (5% annual increase).

This was in contrast to the 44% and 29% mean annual percentage increment of

coniferous tree volume observed during this same period among the manually and cut-

stump treated plantations, respectively. This suggests that the coniferous tree layer did

benefit from the vegetation management treatments in terms of an increased rate of

growth.

Differences in coniferous tree growth within treated plantations may increase relative to

that of the control with time. Simard (1996a) reported that responses of Douglas-fir

height growth to vegetation management treatments were not fully expressed even nine

years after treatment. Also, a more sensitive measure to determine the treatment effects

on crop tree growth may have been possible if the sampling method had included

measurements of stem diameter. Simard (1990, 1996a) reported that stem diameter was

the most sensitive measure of early response of lodgepole pine to release from competing

vegetation.

The slight decrease in coniferous tree volume observed in the first post-treatment year

among the manually and cut-stump treated plantations can likely be attributed to a

phenomenon known as "thinning shock" (Reukema, 1964; Brix, 1981; Harrington and

Reukema, 1983; Maguire, 1983). Thinning shock, of course, does not actually reduce the

stem volume of an affected tree, but rather it temporarily reduces the rate of tree growth

due to increased exposure (e.g., sunscald) and increased physical damage (e.g., wind,

40

snow, and ice) (Harrington and Reukema, 1983). The decrease in coniferous tree cover

observed within the treated plantations during the first post-treatment year was consistent

with the effects of thinning shock.

In addition to the short-term negative effects associated with thinning shock, vegetation

management treatments may have a more serious long-term effect on a stand's

productivity. In deciduous/coniferous mixedwood stands, the potential for deciduous

trees to function as nurse trees for a coniferous crop suggests that a stand may experience

a decrease in production following treatments that suppress the deciduous tree layer (Man

and Lieffers, 1999). Several studies have indicated that deciduous trees can have positive

effects on coniferous tree growth by improving nutrient cycling, interplant transfer of

nutrients and carbon through mycorrhizae, decreasing competition, reducing pest attack,

and by offering physical protection from snow and wind (Navratil et al., 1991; Lieffers

and Beck, 1994, Man and Lieffers, 1999). These studies have all been conducted within

boreal forests and it is not clear how well their results apply to the milder conditions of

the Interior Cedar-Hemlock biogeoclimatic zone. It is clear, however, that no vegetation

management prescription should be chosen without considering the natural ecology of the

area and the non-timber resource values. In addition, the bias towards coniferous

production is not always appropriate with the growing importance of hardwoods in

today's wood products (Man and Lieffers, 1999).

41

Species Richness

Because differences in herb, shrub, and tree species richness observed among control and

treatment plantations were not significant, and time trends were similar among all

plantations, the observed fluctuations can be assumed to be associated with naturally

occurring environmental and successional events. Other studies have also reported no

significant differences in species richness among control plantations and these treated

with manual methods or glyphosate applications (Boyd et al, 1995; Simard and

Heineman, 1996a; 19966; Sullivan etal, 1996; and others).

The sampling method used during this study recorded the presence of a plant, no matter

how small a plant was. Even a plant covering less than 1% was recorded as having trace

coverage within a given sample plot. With this fine-filter approach for vegetation

sampling, a common scenario might be for a small and/or rare species to be sampled

during year 1 and 3, but not during year 2, due to mortality and later germination of that

particular species. This does not imply that this species was necessarily lost from the

site, nor should the reappearance of this species the following year suggest that this

species was gained. Therefore, it was difficult to comment on which species were lost or

gained as changes in species composition, especially for the herb and shrub layers, were

often only temporary.

42

Species Turnover

Herb Layer

The similarity in herb species turnover among control and treatment plantations, as well

as similar trends over time, suggested that herb turnover was not affected by treatments

and that changes may have been the result of naturally occurring successional events.

The number of herb species gained was always greater than species lost during the first

two or three years of the study within control and treatment plantations. The contrary

was true for the final years of the study, when more herb species were lost than gained.

This suggested that the five-year sampling period of this study may have captured a

transition period for the herb layer, perhaps changing from a community of shade-

intolerant to more shade-tolerant herbs in association with increased canopy closure of

the tree layer. The greater number of species gained early in the study may have

represented the last wave of shade-intolerant herbs invading the sites while light

conditions still permitted germination, and the greater number of species lost during the

final years of this study may have represented the loss of shade-intolerant species as light

conditions declined.

The peak in herb species richness observed midway through this study further supported

the hypothesis that the peak in herb species turnover may have been associated with a

period of overlapping herb assemblages (one shade-intolerant and the other more shade-

tolerant), as one would expect greater species richness during such transitions, before

species are lost. Schoonmaker and McKee (1988) reported similar findings during such

transitional periods in a study of secondary succession within coniferous forests of the

43

western Cascade Mountains. They described the transition period from one plant

assemblage to another as a time of unresolved competition.

Unfortunately, I can only speculate as to the cause of the changes in herb species turnover

during this study. Moreover, the lack of statistical difference, within and between years,

suggested that the observed changes in herb species turnover may have been the result of

random variation and not that of the suggested successional progression of plant species.

An analysis that groups herb species into classes of shade-tolerance, as well as a

sampling method that specifically measures canopy closure and light conditions beneath

the forest canopy, would be important for testing this hypothesis (Halpern and Spies,

1995).

Shrub Layer

The similarity in shrub species turnover among control and treatment plantations

suggested that shrub turnover was not affected by treatments and that changes may have

been the result of naturally occurring successional events. However, a weak trend of

increasing shrub turnover within control plantations and a decrease in shrub turnover

within the treated plantations over the 5-year period of this study suggested a possible

treatment effect. Number of species gained and lost within manually and cut-stump

treated plantations remained relatively low and constant throughout the study (average of

0 to 3 species lost and gained within consecutive year intervals). Whereas control

plantations experienced a slightly greater and more variable gain and loss of species

(average of 1 to 4 species lost and gained within consecutive year intervals). Perhaps the

44

treatments have maintained a more stable shrub community by delaying the canopy

closure of the tree layer. The contrary being the case for the control plantations which

may have experienced a more rapid change in shrub community due to the decreasing

light conditions associated with the closing canopy of the tree layer. The decrease in

shrub species turnover observed within both the manual and cut-stump treatments during

the final two years of this study suggested that the delay in canopy closure was only

temporary and that the shrub community within the treated plantations may have begun

to change at this point. As with the herb layer, an analysis that groups shrub species into

classes of shade-tolerance, as well as a sampling method that specifically measures

canopy closure and light conditions beneath the forest canopy, would have been useful

for testing this hypothesis.

Tree Layer

The lack of difference in tree species turnover among control and treatment plantations,

as well as among years, suggested that the treatments did not affect tree species turnover

and that observed changes were likely the result of naturally occurring successional

events.

Species Diversity

Two diversity indices were calculated (Simpson's and Shannon-Wiener) because of the

different interpretations possible from each type of index. However, because time trends

were similar for both indices within the herb, shrub, and tree layers (Figure 5),

45

interpretations of the two diversity indices would also be similar. Therefore, to avoid

redundancy, both Simpson's and Shannon-Wiener indices are discussed together.

When interpreting species diversity results, it is important to consider the two

components that define diversity: 1) the number of species sampled, or species richness,

and 2) the frequency distribution, or relative abundance, of these species. For this study,

crown volume index was used to calculate the proportions of species sampled within an

area. Therefore, crown volume index is useful for inferring information about the

proportion component of a diversity index.

Herb and Shrub Layers

The lack of statistical difference and similar time trends observed among the treatment

and control groups, pre- and post-treatment, suggested that the fluctuations in species

diversity within the herb and shrub layers may be the result of naturally occurring

successional events, and not to the treatments.

Herb species diversity began to decline in 1994, even though this was the year that

experienced the greatest species richness of any of the sample years. However, when

combined with the fact that the greatest herb volume of any of the sample years was also

observed during 1994 (Figure 2), it follows that a few herb species thrived (causing the

increased herb volume) and dominated the herb layer during this time, causing the

observed decline in herb species diversity. Domination was observed to increase from

46

the first to second post-treatment years for several herb species within control and

treatment plantations. In particular, during 1993, the combined crown volumes of

fireweed and grass made up 44, 62, and 63% of the total crown volume of the common

herbs (Table 3) within control, manual, and cut-stump treated plantations, respectively.

In the following year (1994), these same two herbs had increased in dominance and made

up 57, 80, and 67% of the total crown volume of the common herbs within control,

manual, and cut-stump treated plantations, respectively.

The shrub layer species diversity gradually declined from the pre-treatment year (1992)

to the end of the study (1996) among the control, manually, and cut-stump treated

plantations (Figure 5). Although shrub species richness was relatively constant

throughout all five sample years, shrub volume generally increased during the four post-

treatment years. This inverse relationship between shrub species diversity and shrub

volume suggested that a few shrubs became increasingly dominant, causing the observed

decline in diversity, as was concluded for the herb layer. Domination was observed to

increase from the first to second post-treatment years for several shrub species within

control and treatment plantations. In particular, during the first post-treatment year

(1993), the combined crown volumes of thimbleberry, bitter cherry, and willow made up

77, 83, and 63% of the total crown volume of the common shrubs (Table 3) within

control, manual, and cut-stump treated plantations, respectively. By the final year of

sampling (1996), these same three shrub species had increased in dominance so that their

combined volumes made up 86, 91, and 70% of the total crown volume of the common

shrubs within control, manual, and cut-stump treated plantations, respectively.

47

Tree Layer

The manual treatment increased tree species diversity while the cut-stump treatment

resulted in a decrease in diversity from the pre-treatment to post-treatment year period.

Tree species diversity within the control plantations remained relatively unchanged

during this same period. The different treatment effects on tree species diversity between

these two years (1992 and 1993) accounts for the significant Time X Treatment

interaction observed in Table 2, and was confirmed by the nonsignificant interaction

(F 6 j, 8 = 0.68; P = 0.66 and F 6 > i 8 = 0.87; P = 0.53 for Simpson's and Shannon-Wiener

indices, respectively) that resulted from a RM-ANOVA without the pre-treatment year.

Once again, the two components of diversity (number of species (species richness) and

relative proportions of these species (inferred via crown volume index)) must be

investigated to interpret and explain the observed response in tree species diversity.

Because tree species richness was relatively constant and coniferous tree volume changed

little in the first post-treatment year, the dramatic changes in deciduous tree volume were

suspected to be the cause of changes in tree species diversity within the manual and cut-

stump treatments. The dramatic reduction of deciduous tree volume observed within the

cut-stump treated plantations (95% decrease relative to pre-treatment year), while not

eliminating any deciduous species, increased the relative dominance of the coniferous

trees. Of the total crown volume index for the tree layer, coniferous trees made up 50%

during the pre-treatment year. During the first post-treatment year, the proportion of

48

coniferous trees had increased to 94% within plantations receiving the cut-stump

treatment. Moreover, during this first post-treatment year, 88% of the total tree crown

volume was made up of only 4 of the 12 tree species found within the cut-stump treated

plantations, all coniferous (Douglas-fir, western redcedar, lodgepole pine, and western

hemlock). Conifer dominance is, therefore, the most likely cause for the decline in tree

species diversity during the first post-treatment year.

Manually treated plantations also had the deciduous tree layer volume significantly

reduced (87% decrease relative to pre-treatment year). During the pre-treatment year,

76% of the total tree volume was made up of just 3 of the 11 tree species found within the

manually treated plantations, all deciduous (paper birch, black cottonwood, and trembling

aspen). During the year following treatment, these three deciduous tree species

comprised only 37% of the total tree volume. The removal of a substantial proportion of

this deciduous tree layer resulted in a more evenly distributed, and therefore, more

diverse tree layer during the first post-treatment year within manually treated plantations.

Pielou (19666) described this type of increase in species diversity (i.e., one caused by the

thinning of a dominant tree layer) as resulting from an increase in pattern-diversity.

Structural Diversity

Herb Layer

The similar structural diversity (both Simpson's and Shannon-Wiener) of the herb layer

observed among control and treatment plantations, pre- and post-treatment, suggested

49

that changes in structural diversity may have been caused by naturally occurring

successional events, and not by the treatments. The gradual decline in structural diversity

of the herb layer within control and treatment plantations was likely correlated with

increasing canopy closure of the tree layer. By removing a substantial portion of the tree

canopy, the time to canopy closure was delayed within plantations that received

treatments. The observed delay for the herb layer to reach peak structural diversity

within the manual and cut-stump treated plantations further suggested that structural

diversity of the herb layer was related to the overstory canopy closure.

I predicted that the treatments should not significantly alter the structural diversity of the

herb layer. This was because of the annual nature of herbs which allow this plant layer to

respond quickly to changing environmental conditions, growing into many of the same

height classes as it had the year before. In contrast, perennial plants such as shrubs,

because of their interdependence on the previous year's growth, were predicted to take a

little longer, as a group, to respond to new environmental conditions.

Shrub Layer

The lack of statistical difference among the control and treatment plantations for

structural diversity of the shrub layer suggested that the minor fluctuations in diversity

may have been caused by naturally occurring successional events. However, the slight

decrease and increase in structural diversity observed within treated and control

plantations, repectively, during the first post-treatment year suggested a possible

treatment effect. By felling a substantial portion of the deciduous tree layer during

50

treatment, it is speculated that the obstruction and physical damage caused to the shrub

layer (particularly the taller height classes) may have temporarily simplified the structural

diversity of this layer.

Tree Layer

As a stand of trees ages, shading created by increasing canopy closure begins to thin out

and simplify the understory height classes (Pielou, 19666). By removing a substantial

portion of the canopy, the manual and cut-stump treatments had effectively delayed the

onset of canopy closure within the treated plantations. The result was that the plantations

receiving treatments had greater structural diversity than the control plantations, although

this difference was not significant. The control plantations moved quickest towards

canopy closure, and as such, decreased in structural diversity faster than either of the two

treatment groups.

Total (herb, shrub, and tree layers combined)

Because overstory shading, to a large extent, governs what can grow in the understory,

canopy closure is a driving force that will largely determine the total (herb, shrub, and

tree layers combined) structural diversity of a stand. Therefore, as the control plantations

grew unimpeded towards canopy closure, there was a steady decline in structural

diversity throughout all five years of this study. Treatments, on the other hand,

temporarily opened the overstory canopy, and consequently the total structural diversity,

although decreasing through the post-treatment years, was statistically more structurally

diverse than that of the control in 1994.

51

Conclusions

This study suggested that conifer release from deciduous trees (capable of stump sprouts)

was best achieved with a cut-stump application of herbicide, such as glyphosate.

Methods that do not employ some means of impeding stump sprouts, such as the manual

cutting method, are not likely to maintain suppression of deciduous trees for extended

periods following treatment.

Although the cut-stump herbicide application of glyphosate significantly reduced

deciduous tree volume for four post-treatment years, this treatment had little effect on

species richness, species diversity, and structural diversity of the plant community (herb,

shrub, and tree layers). In fact, reduced dominance of the deciduous tree layer and

opening of the tree canopy created by both treatments appeared to increase the total

structural diversity (all layers combined) during the post-treatment years, relative to that

of the control.

This study appears to be the first to analyze a plant community's response to alternative

vegetation management treatments applied to young mixed-conifer plantations. Results

suggested that the species-specific approach of the cut-stump herbicide treatment should

achieve its objective of conifer release without adversely affecting the species richness,

species diversity, or structural diversity of the plant community, while suppressing the

targeted deciduous tree layer volume.

52

Literature Cited

Adler, G.H. 1987. Influence of habitat structure on demography of two rodent species in

eastern Massachusetts. Can. J. Zool. 65: 903-912.

Anthony, R.G. and M.L. Morrison. 1985. Influence of glyphosate herbicide on small

mammal populations in western Oregon. Northwest Sci. 59:159-168.

B.C. Ministry of Forests. 1995. Forest practices code of British Columbia: Biodiversity

guidebook. Ministry of Forests and Ministry of Environment. 99 p.

Balda, R.P. 1969. Foliage use by birds of the oak-pine juniper woodland and ponderosa

pine forest in southeastern Arizona. Condor. 7 1 : 399-412.

Berry, A.B. 1982. Response of suppressed conifer seedlings to release from aspen-pine

overstory. For. Chron. 58: 91-92.

Boyd, R.S., J.D. Freeman, J.H. Miller, and M.B. Edwards. 1995. Forest herbicide

influences on floristic diversity seven years after broadcast pine release treatments in

central Georgia, USA. New Forests 10. 1:17-37.

Brand, D.G. 1992. The use of vegetation management in Canadian forest regeneration

programs. Aspects of Appl. Bio. 29:113-124.

Brix, H. 1981. Effects of thinning and nitrogen fertilization on branch and foliage

production in Douglas-fir. Can. J. For. Res. 1 1 : 502-511.

Cambell, D.L., J. Evans, G.D. Lindsey, and W.E. Dusenberry. 1981. Acceptance by

black-tailed deer of foliage treated with herbicides. USDA For. Serv., Pacific

Northwest Forest and Range Experiment Station; Portland, OR: USDI, Fish and

Wildlife Service, Forest-Animal Damage Control Research Project. Olympia, WA.

Res. Pap. PNW-290. 31 p.

Campbell, R.A. 1990. Herbicide use for forest management in Canada: where we are

and where we are going? For. Chron. 66: 355-360.

Christensen, P. 1984. Review of Danish results from chemical/mechanical control of

deciduous vegetation. Aspects of Appl. Bio. 5:135-142.

Clough, G.C. 1987. Relations of small mammals to forest management in northern

Maine. Can. Field-Nat. 101:40-48.

Crossley, D. I. 1976. Growth response of spruce and fir to release from suppression.

For. Chron. 52: 189-193.

D'Anieri, P., Leslie, D.M., Jr., and M.L. McCormack, Jr. 1987. Small mammals in

glyphosate-treated clearcuts in northern Maine. Can. Field-Nat. 101: 547-550.

Freedman, B. 1991. Controversy over the use of herbicides in forestry, with particular

reference to glyphosate usage. J. Environ. Sci. Health. C8: 277-286.

Freedman, B., R. Morash, and D. MacKinnon. 1993. Short-term changes in vegetation

after the silvicultural spraying of glyphosate herbicide onto regenerating clearcuts in

Nova Scotia, Canada. Can. J. For. Res. 23:2300-2311.

Glover, P. 1994. Reducing forest herbicide use in Northwestern British Columbia:

Public pressure gets results. J. Pest. Reform. 14: 2-4.

Halpern C B . and T.A. Spies. 1995. Plant species diversity in natural and managed

forests of the Pacific Northwest. Ecological Applications. 5:913-934.

Harney, B.A. and R.D. Dueser. 1987. Vertical stratification of activity of two

Peromyscus species: an experimental analysis. Ecology. 68: 1084-1091.

54

Harrington, C A . and D.L. Reukema. 1983. Initial shock and long-term stand

development following thinning in a Douglas-fir plantation. For. Sci. 29: 33-46.

Hart, D. and P.G. Comeau. 1992. Manual brushing for forest vegetation management in

British Columbia: a review of current knowledge and information needs. B.C. Min.

For., Victoria, B.C., Land Manage. Handb. No. 7.

Hitchcock, C.L. and A. Cronquist. 1973. Flora of the Pacific Northwest. University of

Washington Press, Seattle.

Hunter, M.L., Jr. 1990. Vertical structure (Chapter 11). In Wildlife, forests, and

forestry: principles of managing forest for biological diversity. Prentice Hall,

Englewood Cliffs, N.J. pp 181-199.

Hurlbert, S. H. 1984. Pseudoreplication and the design of ecological field experiments.

Ecol. Monogr. 54: 187-211

Johansson, T. 1985. Herbicide injections into stumps of aspen and birch to prevent

regrowth. Weed Res. 25: 39-45.

Jones, R. and J.M. Forbes. 1984. A note on effects of glyphosate and quinine on the

palatability of hay for sheep. Animal Production. 38:301-303.

Ketcheson, M.V., Braumandl, T.F., Meidinger, D., Utzig, G., Demarchi, D.A., and

B.M.Wikeem, 1991. Interior Cedar-Hemlock zone. In Ecosystems of British

Columbia. Edited by D. Meidinger and J. Pojar. B.C. Ministry of Forests, Victroria.

pp 167-181.

Krebs, C.J. 1989. Ecological Methodology. Harper and Row, New York, New York,

USA.

Kuehl, R.O. 1994. Repeated measures designs (Chapter 15). In Statistical principles of

research design and analysis. Duxbury Press, Belmont, California, pp 499-528.

Lautenschlager, R.A. 1986. Forestry, herbicides, and wildlife. In Is good forestry good

wildlife management? Edited by J.A. Bissonette. Maine Agric. Stn. Misc. Publ. 689.

pp. 299-308.

Lautenschlager, R.A. 1993. Response of wildlife to forest herbicide applications in

northern coniferous ecosystems. Can. J. For. Res. 23: 2286-2299.

Lavigne, M.B. 1988. Growth and net assimilation rates in thinned and unthinned stands

of balsam fir. Can. J. For. Res. 18: 1205-1210.

Lieffers V.J. and J.A. Beck Jr. 1994. A semi-natural approach to mixedwood

management in the prairie provinces. For. Chron. 70: 260-264.

Littel, R.C. 1989. Statistical analysis of experiments with repeated measures.

HortScience. 24: 36-40.

Lloyd, D.A., Angove, K., Hope, G., and C. Thompson, 1990. A guide for site

identification and interpretation of the Kamloops Forest Region. Vols. 1 and 2. B.C.

Ministry of Forests, Victoria. Land Manage. Handb. 23.

Lund-Haie, K. 1984. Growth response of Norway spruce (Picea abies L.) to different

vegetation management programmes - preliminary results. Aspects of Appl. Bio. 5:

127-133.

MacArthur, R.H. 1964. Environmental factors affecting bird species diversity. Am. Nat.

98: 387-397.

MacArthur, R.H. 1965. Patterns of species diversity. Biol. Rev. 40:510-533.

MacArthur, R.H. and J.W. MacArthur. 1961. On bird species diversity. Ecology. 42:

594-598.

MacKinnon, D.S. and B. Freedman. 1993. Effects of silviculture use of the herbicide

glyphosate on breeding birds of regenerating clearcuts in Nova Scotia, Canada. J.

Appl. Ecol. 30 395-406.

MacLean, D.A. and M.G. Morgan. 1983. Long-term growth and yield response of

young fir to mechanical and chemical release from shrub competition. For. Chron.

59: 177-183.

Maguire, D.A. 1983. Suppressed crown expansion and increased bud density after

precommercial thinning in California Douglas-fir. J. For. Res. 13: 1246-1248.

Man, R. and V.J. Lieffers. 1999. Are mixtures of aspen and white spruce more

productive than single species stands? For. Chron. 75: 505-513.

Marrs, R.H. 1985. Birch control by the treatment of cut stumps with herbicides. Arbor.

J. 9: 173-182.

McDonald, P.M. and S.R. Radosevitch. 1992. General principles of forest vegetation

management. In Silvicultural approaches to animal damage management in Pacific

Northwest forests. Technical Editor: H.C. Black. USDA For. Serv. Gen. Tech. Rep.

PNW-GTR-287. pp 67-91.

McGee, A.B. and E. Levy. 1988. Herbicide use in forestry: communication and

information gaps. J. Environ. Manage. 26: 111-126.

Mitchell, W. 1990. Vegetation management and public concerns in British Columbia. In

Vegetation management: An integrated approach - Proceeding of the fourth annual

57

vegetation management workshop, Nov. 14-16, 1989. Vancouver, B.C. Compiled by

E. Hamilton. Ministry of Forests, Research Branch, pp. 8-9.

Morrison, M.L. and E.C. Meslow. 1983. Impacts of forest herbicides on wildlife:

toxicity and habitat alteration. Trans. N . Am. Wildl. Nat. Resour. Conf. 48: 175-185.

Navratil, S., K. Branter and J. Zasada. 1991. Regeneration in the mixedwoods. pages

32-48. In A. Shortreid, Ed. Northern mixedwood '89. Proc. Symp., Sept. 12-14, 1989.

Fort St. John, B.C. Pacific Forestry Centre. Victoria, B.C. FRDA Rep. 164.

Newton, M. and P.G. Comeau. 1990. Control of competing vegetation, irc Regenerating

British Columbia's forests. Edited by D.P. Lavender, R. Parish, C M . Johnson, G.

Montgomery, A. Vyse, R.A. Willis, and D. Winston. University of British Columbia

Press, Vancouver, pp. 256-265.

Newton, M. , Howard, K.M., Kelpsas, B.R., Danhaus, R., Lottman, C M . , and S.

Dubelman. 1984. Fate of glyphosate in an Oregon forest ecosystem. J. Agric. Food

Chem. 32: 1144-1151.

Peet, R.K. 1974. The measurement of species diversity. Annu. Rev. Ecol. Syst. 5: 285-

307.

Pielou, E.C. 1966a. The measurement of species diversity in different types of

biological collections. J. Theor. Biol. 13:131-144

Pielou, E.C. 19666. Species-diversity and pattern-diversity in the study of ecologial

succession. J. Theor. Biol. 10: 370-383.

Pollack, J.C, P. LePage, and F. van Thienen. 1990. Some effects of different forest

herbicides on upland Salix spp. Can. J. For. Res. 20: 1271-1276.

Reukema, D.L. 1964. Crown expansion and stem radial growth of Douglas-fir as

influenced by release. For. Sci. 10:192-199.

Ritchie, R. T., and T. P. Sullivan. 1989. Monitoring methodology for assessing the

impact of forest herbicide use on small mammal populations in British Columbia.

FRDA report no. 081.

Rosenthal, R. and R.L. Rosnow. 1985. Contrast analysis: focused comparisons in the

analysis of variance. Cambridge University Press. London.

Runciman, J. B. and T. P. Sullivan. 1996. Influence of alternative conifer release

treatments on habitat structure and small mammal populations in south central British

Columbia. Can. J. For. Res. 26: 2023-2034.

Santillo, D.J., Leslie, D.M., Jr., and P.W. Brown. 1989. Response of small mammals

and habitat to glyphosate application on clearcuts. J. Wild. Manage. 53: 164-172.

Schoonmaker, P. and A. McKee. 1988. Species composition and diversity during

secondary succession of coniferous forests in the western Cascade Mountains of

Oregon. For. Sci. 34: 960-979.

Simard, S.W. 1990. A retrospective study of competition between paper birch and

planted Douglas-fir. Can. For. Serv. and B.C. Min. For., Victoria, B.C. FRDA Rep.

No. 147.

Simard, S.W. and J. Heineman. 1996a. Nine-year response of Douglas-fir and the

mixed-hardwood-shrub complex to chemical and manual release treatments on an

ICHmw2 site near Salmon Arm. Can. For. Serv. and B.C. Min. For., Victoria, B.C.,

FRDA Rep. No. 257.

59

Simard, S.W. and J. Heineman. 19966. Nine-year response of lodgrepole pine and the

dry alder complex to chemical and manual release treatments on an ICHmkl site near

Kelowna. Can. For. Serv. and B.C. Min. For., Victoria, B.C., FRDA Rep. No. 259.

Simard, S.W. and J. Heineman. 1996c. Nine-year response of Englemann spruce and the

willow complex to chemical and manual treatments on an ICHmw2 site near Vernon.

Kelowna. Can. For. Serv. and B.C. Min. For., Victoria, B.C., FRDA Rep. No. 258.

Simpson, E.H. 1949. Measurement of diversity. Nature. 163:688.

SPSS Institute Inc. 1997. Statistical Programs for the Social Sciences. Chicogo, IL.

Stickney, P. F. 1980. Data base for post-fire succession, first 6 to 9 years, in Montana

Larch-fir forests. USDA Forest Service General Technical Report, LNT-62, Ogden,

UT: U.S. Department of Agriculture, Forest Service, Intermountain Research Station.

Stickney, P. F. 1985. Data base for early post-fire succession on the Sundance Burn,

northern Idaho. USDA Forest Service General Technical Report, INT-189, Ogden,

UT: U.S. Department of Agriculture, Forest Service, Intermountain Research Station.

Sullivan, T.P. 1990a. Influence of forest herbicide on deer mice and Oregon vole

population dynamics. J. Wildl. Manage. 54: 566-576.

Sullivan, T.P. 19906. Demographic responses of small mammal populations to a

herbicide application in coastal coniferous forest: population density and resiliency.

Can. J. Zool. 68: 874-883.

Sullivan, T.P. 1994. Influence of herbicide-induced habitat alteration on vegetation and

snowshoe hare populations in sub-boreal spruce forest. J. Appl. Ecol. 31: 717-730.

Sullivan, T.P. and D.S. Sullivan. 1979. The effects of glyphosate herbicide on food

preference and consumption in black-tailed deer. Can. J. Zool. 57: 1406-1412.

60

Sullivan, T.P. and D.S., Sullivan. 1981. Response of a deer mouse population to a forest

herbicide application: reproduction, growth, and survival. Can. J. Zool. 59: 1148-

1154.

Sullivan T.P. and D.S., Sullivan. 1982. Responses of small mammal populations to a

forest herbicide application in a 20-year-old conifer plantation. J. Appl. Ecol. 19: 95-

106.

Sullivan, T.P., D.S. Sullivan, R.A. Lautenschlager, and R.G. Wagner. 1997. Long-term

influence of glyphosate herbicide on demography and diversity of small mammal

communities in coastal coniferous forest. Northwest Sci. 71: 6-17.

Sullivan, T.P., R.A. Lautenschlager, and R.G. Wagner. 1996. Influence of glyphosate on

vegetation dynamics in different successional stages of sub-boreal spruce forest.

Weed Tech. 10:439-446.

Sullivan, T.P., R.G. Wagner, D.G. Pitt, R.A. Lautenschlager, and D.G. Chen. 1998.

Changes in diversity of plant and small mammal communities after herbicide

application in sub-boreal spruce forest. Can. J. For. Res. 28:168-177

Sutton, S.L. and P.J. Hudson. 1980. The vertical distribution of small flying insects in

lowland rainforests of Zaire. Zool. J. Linnaean Soc. 68:111-123.

Tomkins, D.J. and W.F. Grant. 1977. Effects of herbicides on species diversity of two

plant communities. Ecology. 58: 398-406.

Turpin, T.C. 1990. Vegetation management and public issues in the Pacific Northwest.

In Vegetation management: An integrated approach - Proceeding of the fourth annual

vegetation management workshop, Nov. 14-16,1989. Vancouver, B.C. Compiled by

E.Hamilton. Ministry of Forests, Research Branch, pp 4-7.

61

Wagner, R. G. 1993. Research directions to advance forest vegetation management in

North America. Can. J. For. Res. 23: 2317-2327

Wall, R.E. 1990. The fungus Chrondrostereum purpureum as a silvicide to control

stump sprouting in hardwoods. North. J. Appl. For. 7:17-19.

Walmsley, M. E., G. Utzig, T. Void, and J. Van Barneveld. 1980. Describing

ecosystems in the field. British Columbia Ministry of Environment and Ministry of

Forests, Land Management Report Number 7.

Wang, J.R., S.W. Simard, J.P. Kimmins. 1995. Physiological responses of paper birch to

thinning in British Columbia. For. Ecol. Manag. 73: 177-184.

Whitehead, R.J. and G.J. Harper. 1998. A comparison of four treatments for weeding

Englemann spruce plantations in Interior Cedar Hemlock Zone of British Columbia:

ten years after treatment. Information Report BC-X-379. Canadian Forest Service,

Pacific Forestry Centre, Victoria, BC. 21 p.

Yang, R.C. 1991. Growth of white spruce following release from aspen competition: 35

year results. For. Chron. 67:706-711.

Zar, J.H. 1984. Data transformations (Chapter 14). In Biostatistical analysis. 2 n d ed.

Prentice Hall, Englewood Cliffs, N.J. pp 236-243.

62

Appendix 1

Alphabetical list of all of the herb species study (75 species in total). Nomenclature

that were sampled during the five years of this follows Hitchcock and Cronquist, (1973).

Scientific (Latin) name Common name Achillea millefolium L.

Actea rubra (Ait.) Willd.

Adenocaulon bicolor Hook.

Anaphalis margaritacea L.

Antennaria microphylla Rydb.

Apocynum androsaemifolium L.

Aquilegia formosa Fisch.

Aralia nudicaulis L.

Arnica cordifolia Hook.

Asarum caudatum Lindl.

Aster ciliolatus Lindl.

Aster foliaceus Lindl.

Athyrium filix-femina (L.) Roth.

Calypso bulbosa (L.) Oakes.

Car ex spp. L.

Castilleja miniata Dougl.

Cerastium arvense L.

Chrysanthemum leaucanthemum L.

Cirsium arvense (L.) Scop.

Cirsium vulgare (Savi) Tenore

Clintonia uniflora (Schult.) Kunth.

Cornus canadensis L.

Disporum hookeri (Torr.) Nicholson

Epilobium angustifolium L.

Epilobium minutum Lindl.

Yarrow

Baneberry

Pathfinder

Pearly everlasting

Rosy pussytoes

Spreading dogbane

Red columbine

Wild sarsaparilla

Heart-leaved arnica

Wild ginger

Lindley's aster

Leafy aster

Lady fern

Fairyslipper

Sedge

Common red paintbrush

Field chickweed

Oxeye daisy

Canada thistle

Bull thistle

Quene's cup

Bunchberry

Hooker's fairybells

Fireweed

Small-flowered willowherb

63

Appendix 1 Continued Scientific (Latin) name Common name Epilobium watsonii Barbey Purple-leaved willowherb

Equisetum arvense L. Common horsetail

Fragaria virginiana Duchesne Wild strawberry

Galium triflorum Michx. Sweet-scented bedstraw

Geranium sp. L. Geranium

Geum macrophullum Willd. Large-leaved avens

Goody era oblongifolia Raf. Rattlesnake plantain

Gramineae spp. Grass

Gymnocarpium dryopteris (L.) Newm. Oak fern

Habenaria unalescensis (Spreng.) Wats. Alaska rein-orchid

Hieracium albiflorum Hook. White hawkweed

Hieracium aurantiacum L. Orange hawkweed

Hieracium cynoglossoides Arv.-Touv. Hounds-tongue hawkweed

Hieracium gracile Hook. Slender hawkweed

Hieracium umbellatum L. Narrow-leaved hawkweed

Lactuca canadensis L. Canadian wild lettuce

Lactuca muralis (L.) Fresen. Wall lettuce

Lactucapulchella (Pursh) DC. Blue lettuce

Lathyrus ochroleucus Hook. Creamy peavine

Lilium columbianum Hanson Tiger lily

Lupinus latifolius var. subalpinus (Piper &

Robins) Smith .TA.1. C 1 U U 1 1 I C

Medicago sativa L. Alfalfa

Melilotus alba Desr. White sweet-clover

Melilotus officinalis (L.) Lam. Yellow sweet-clover

Mentha arvensis L. Field mint

64

Appendix 1 Continued Scientific (Latin) name Common name Mitella nuda L.

Osmorhiza chilensis H. & A.

Osmorhiza purpurea (Coult. & Rose)

Suksd.

Prunella vulgaris L.

Pteridium aquilinum (L.) Kuhn.

Pyrola asarifolia Michx.

Pyrola picta Smith

Pyrola secunda L.

Ranunculus sp. L.

Saxifraga occidentalis Wats.

Smilacina racemosa (L.) Desf.

Smilacina stellata (L.) Desf.

Sonchus arvensis L.

Streptopus amplexifolius (L.) DC.

Streptopus roseus Michx.

Streptopus streptopoides (Ledeb.) Frye &

Rigg Taraxacum officinale Weber

Tiarella trifoliata var. trifoliata L.

Tiarella trifoliata var. unifoliata (Hook.)

Kurtz.

Tragopogon dubius Scop.

Trifolium pratense L.

Trifolium repens L.

Verbascum thapsus L.

Vicia americana Muhl.

Viola spp. (L.)

Common mitrewort

Mountain sweet-cicely

Purple sweet-cicely

Self-heal

Bracken

Pink wintergreen

White-viened wintergreen

One-sided wintergreen

Buttercup

Western saxifrage

False Solomon' s-seal

Star-flowered false Solomon's-seal

Perennial sow-thistle

Clasping twisted stalk

Rosy twisted stalk

Small twisted stalk

Common dandelion Three-leaved foamflower

One-leaved foamflower

Yellow salsify

Red clover

White clover

Great Mullein

American vetch

Violet

65

Appendix 2

Alphabetical list of all of the shrub species that were sampled during the five years of this study (37 species in total). Nomenclature follows Hitchcock and Cronquist, (1973). Scientific (Latin) name Common name Acer glabrum var. douglasii (Hook.)

Dippel Alnus incana (L.) Moench.

Amelanchier alnifolia Nutt.

Arctostaphylos uva-ursi (L.) Spreng.

Berberis aquifolium Pursh

Ceanothus velutinus Dougl.

Chimaphila umbellata (L.) Bart.

Cornus stolonifera Michx.

Gaultheria ovatifolia Gray

Holodiscus discolor (Pursh) Maxim

Linnaea borealis (L.)

Lonicera ciliosa (Pursh) DC.

Lonicera involucrata (Rich.) Banks

Lonicera utahensis Wats.

Menziesia ferruginea Smith

Pachistima myrsinites (Pursh) Raf.

Prunus emarginata (Dougl.) Walp.

Rhamnus purshiana DC.

Ribes lacustre (Pers.) Poir.

Ribes viscosissimum Pursh.

Rosa acicularis Lindl.

Rosa gymnocarpa Nutt.

Rubus idaeus L.

Rubus leucodermis Dougl.

Rubus parviflorus Nutt.

Douglas maple

Speckled alder

Saskatoon

Kinnikinnick

Tall Oregon-grape

Snowbrush

Prince's-pine

Red-osier dogwood

Oregon wintergreen

Ocean spray

Twinflower

Orange honeysuckle

Black twinberry

Utah honeysuckle

False azalea

Falsebox

Bitter cherry

Cascara

Black gooseberry

Sticky current

Prickly rose

Baldhip rose

Red raspberry

Black raspberry

Thimbleberry

66

Appendix 2 Continued Scientific (Latin) name Common name Rubus pedatus J. E . Smith Five-leaved bramble

Rubus pubescens Raf. Trailing raspberry

Salix spp. L. Willow

Sambucus racemosa L. Red elderberry

Sheperdia canadensis (L.) Nutt. Soopolallie

Sorbus sitchensis Roemer Sitka mountain-ash

Spiraea betulifolia Pall. Birch-leaved spirea

Symphoricarpos albus (L.) Blake Common snowberry

Vaccinium caespitosum Michx. Dwarf blueberry

Vaccinium membranaceum Dougl. Black huckleberry

Vaccinium ovalifolium Smith Oval-leaved blueberry

Vaccinium parvifolium Smith Red huckleberry

67

Appendix 3

Alphabetical list of all of the tree species that were sampled during the five years of this study (12 species in total). Nomenclature follows Hitchcock and Cronquist (1973). Scientific (Latin) name Common name Abies lasiocarpa (Hook.) Nutt.

Betula papyrifera Marsh.

Larix occidentalis Nutt. Picea engelmannii Parry X Picea glauca (Moench) Voss Pinus contorta Dougl. ex Loud. var. latifolia Engelm. Pinus monticola Dougl.

Populus tichocarpa T. & G.

Populus tremuloides Michx. Pseudotsuga menziesii var. glauca (Beissin.) Franco Taxus brevifolia Nutt.

Thuja plicata Donn ex D. Don

Tsuga heterophylla (Raf.) Sarg.

Subalpine fir

Paper birch

Western larch

Hybrid spruce

Lodgepole pine

Western white pine

Black cottonwood

Trembling aspen

Interior Douglas-fir

Western yew

Western redcedar

Western hemlock

68