Lakehead University
Knowledge Commons,http://knowledgecommons.lakeheadu.ca
Research and scholarly works Faculty of Natural Resources Management
2011
Benefit–cost analysis of vegetation
management alternatives: an Ontario
case study
Luckai, Nancy J
The Forestry Chronicle, 2011, 87(2): 260-273, http://dx.doi.org/10.5558/tfc2011-013
http://dx.doi.org/10.5558/tfc2011-013
Downloaded from Lakehead University, KnowledgeCommons
260
ma r s/av r il
2011, v o l . 87, N
o
2 — Th e For esTr y Ch r oNiCl e
Benefit–cost analysis of vegetation management alternatives:
An Ontario case study
by Krish Homagain
1,2
, Chander K. Shahi
1
, Nancy J. Luckai
1
, Mathew Leitch
1
and F. Wayne Bell
3
ABSTRACT
Vegetation management practices are an integral component of forest management. In this paper, we report results of
stand-level benefit–cost analyses of 12 vegetation management treatments applied at six study sites in northern Ontario.
Forest Vegetation Simulator (FVS
Ontario
) was used to project gross total and merchantable volumes to 70 years of age, and
BUCK-2 was used to optimize potential products. Net present value (NPV), benefit–cost ratio (BCR), and internal rate
of return (IRR) were calculated using 2009 constant dollars and variable real discount rates. Aerial herbicide treatments
produced the highest NPV, BCR, and IRR. Internal rates of return of 4.32%, 2.90%, 2.82% and 2.50% for aerial herbicide,
manual brush cutting, ground-applied herbicide, and brush cutting plus herbicide treatments, respectively, indicated that
all of the vegetation management alternatives evaluated are economically viable.
Key words: aerial herbicides, brush saw, forest economics, Forest Vegetation Simulator (FVS
Ontario
), ground herbicides,
internal rate of return, net present value
RÉSUMÉ
Les pratiques de contrôle de la végétation font partie intégrante de l’aménagement forestier. Dans cet article, nous
reportons des résultats obtenus par des analyses de coût-bénéfice effectuées au niveau du peuplement pour 12 traitements
de contrôle de la végétation appliqués sur six sites d’études du nord de l’Ontario. Le Forest Vegetation Simulator
(FVSOntario) a été utilisé pour projeter le volume total et le volume marchand à l’âge de 70 ans et le BUCK-2 a été utilisé
pour optimiser les produits potentiellement obtenus. La valeur actuelle nette (VAN), le ratio coût-bénéfice (RCB) et le
taux de rentabilité interne (TRI) ont été calculés en dollars constants de 2009 et selon des taux variables de profitabilité
réelle. L’épandage aérien de phytocide a généré les VAN, RCB et TRI les plus importants. Les taux de rentabilité interne de
4,32 %, 2,90 %, 2,82 % et de 2,50 % respectivement pour l’épandage aérien, le débroussaillage manuel, l’application
terrestre de phytocide et le débroussaillage manuel suivi de traitements de phytocide, ont démontré que toutes les
alternatives de contrôle de la végétation évaluées sont économiquement viables.
Mots clés : épandage aérien de phytocide, débroussailleuse, économie forestière, Forest Vegetation Simulator
(FVSOntario), épandage terrestre de phytocide, taux de rentabilité interne, valeur actuelle nette
1
Faculty of Natural Resources Management, Lakehead University, 955 Oliver Rd., Thunder Bay, Ontario P7B 5E1.
2
Corresponding author. E-mail: khomagai@lakeheadu.ca
3
Ontario Forest Research Institute, Ontario Ministry of Natural Resources, 1235 Queen Street East, Sault Ste. Marie, Ontario P6A 2E5.
Introduction
Ontario’s forest sector is a key component of the province’s
economy (OMNDMF 2010). Most of Ontario’s productive
forest is in the conifer-dominated boreal region where opti-
mization of growth rates of spruce (Picea) and pine (Pinus)
species is a key forest management objective, but these species
are often out-competed by hardwoods such as poplar (Popu-
lus spp.) (Hearnden et al. 1992). Maintaining overall forest
composition is a legal requirement under Ontario’s Crown
Forest Sustainability Act, which states that “large, healthy,
diverse and productive Crown forests and their associated
ecological processes and biological diversity should be con-
served” (Statutes of Ontario 1995). As a result, forest vegeta-
tion management practices are an integral component of for-
est management.
Forest vegetation management practices help to ensure
initial plantation
4
survival, accelerate growth of targeted
species, and achieve high yields in terms of per unit gross total
volume production (Wagner et al. 2006). Forest vegetation
management practices include several alternatives (Wien-
sczyk et al. 2011, this issue) and results from experimental
studies on growth rates and volume production are highly
variable among these alternative treatments (Comeau et al.
1999; Simard et al. 2001; Heineman et al. 2005; Bell et al.
2011a, this issue). The costs associated with the various treat-
ments also vary greatly (Bell et al. 1997, Dampier et al. 2006).
In a system where both yields and costs vary, an economic
analysis of the efficiency of a silvicultural intervention can
only be evaluated based on long-term stand-level growth
response data and cost information (McKenney et al. 1997).
In a review of the Canadian forest vegetation management
research and practice, Thompson and Pitt (2003) report 1256
scientific publications directly related to forest vegetation
management as of 2002, but only 18 (1.4%) of those include
components of economic analysis of forest vegetation man-
agement treatments, and even fewer are focused on the eco-
nomics associated with releasing boreal conifers. Therefore,
4
Plantation is a forest crop established artificially, either by sowing
or by planting (NRCan 1995)
The Forestry Chronicle Downloaded from pubs.cif-ifc.org by Lakehead University on 10/28/16
For personal use only.
ma r Ch /apr il 2011, v ol . 87, No. 2 — Th e Fo r esTr y Ch r oNiCl e 261
there is a need for stand-level benefit–cost analyses (BCA)
that will help decision-makers choose the best alternatives for
forest vegetation management. BCA is a method of apprais-
ing and evaluating an investment decision that includes iden-
tification, valuation, and comparison of all costs and benefits
during the life of a project (Campbell and Brown 2003). BCA
provides the most comprehensive framework for evaluating
any economic investment, as it estimates values associated
with inputs and outputs for each activity (Nautiyal et al.
2001). Net present value (NPV), benefit–cost ratio (BCR) and
internal rate of return (IRR) are the most commonly used
measures for conducting benefit–cost analysis. NPV
expresses the difference between the discounted present value
of future benefits and the discounted present value of future
costs, whereas BCR is the ratio of discounted present value of
benefits to the discounted present value of costs. A positive
NPV or a BCR greater than one indicates that the activity
being evaluated is economically beneficial. We used both
NPV and BCR because for investment decisions related to
forestry activities, NPV gives the best comparison when silvi-
culture budgets are not limited and BCR gives the best com-
parison when silviculture budgets are limited (Willcocks et al.
1997). However, both NPV and BCR depend on the discount
rate used for the analysis. Therefore, the benefit–cost analysis
is supplemented by finding the IRR, which is the discount
rate when NPV is zero. An IRR greater than the existing mar-
ket interest rates in general indicates a relatively profitable
investment (Campbell and Brown 2003). Such analyses pro-
vide a management tool to evaluate and compare amongst
different release treatments, thereby building a competitive,
knowledge-based forest industry that is sustainable under
increasing global competition.
In this paper, which is one of a series of papers related to
forest vegetation management published in the March/April
2011 issue of The Forestry Chronicle (see Bell et al. 2011b, this
issue), we report the results of stand-level BCA of vegetation
management treatments applied at six sites in northern
Ontario. The specific objectives of the study were to: (i) calcu-
late costs associated with each vegetation management treat-
ment over almost two decades, (ii) estimate projected yield
and value of fibre (timber, pulpwood, and hog fuel) produc-
tion using a simulation and an optimization model, and (iii)
conduct BCA to compare the economic viability of the vege-
tation management treatments.
Methods
The simulations and benefit–cost analyses presented in this
paper are based on data from six Vegetation Management
Alternative Program (VMAP) studies. In brief, yields were
projected beyond the data to age 70 years using Forest Vege-
tation Simulator (FVS
Ontario
) (ESSA 2008). BUCK-2 opti-
mization software (Zakrzewski et al. 2010) was used to deter-
mine what forest products could be produced at 70 years
following treatments. Future benefits were calculated using
current Thunder Bay market prices for pulpwood and hog
fuel
1
, and lumber prices from Random Length price statistics
for 2009 (Random Lengths 2009). A range of real discount
rates (2% to 10%) was used to calculate net present value and
benefit–cost ratio for each vegetation management treat-
ment. In addition, the internal rate of return for each treat-
ment was estimated to compare changes in NPV over differ-
ent discount rates. Details of each stage of analysis are
provided below following descriptions of the studies from
which the data were obtained (for additional details about
study sites, see Bell et al. 2011a).
Study areas and vegetation management treatments
Data collected over 10 to 16 years from six research studies in
northern Ontario were used for the analysis (Table 1). All sites
were clearcut harvested, mechanically site prepared
Table 1. Vegetation management alternative program study areas, treatments, and experimental designs from which data were
obtained for the benefit-cost analysis (adapted from Bell et al. 2011a)
Study site Location Year planted Crop species Year released Release treatments
a
Exp. design (Exp. units)
Bending Lake 48°57
′N 1988 jack pine 1992–93 ASg, BS, CON, CRg RCBD: 4 Blocks (16)
92°02
′ W
Espanola 46°48
′N Block 1: 1989 jack pine 1993 ASg, BBt, BS, CON, RCBD: 3 Blocks (18)
82°11
′W Others: 1991 CRg, MBg,
Fallingsnow 48°08
′N 1987–90 white spruce 1993 ASg, ASt, BS, CON, RCBD: 3 Blocks (21)
89°49
′W CRb, CRg, SIL
Leether Lake 50°36
′N 1988 jack pine, 1993 ASg, BS, CON, CRg CRD: 3 Blocks (12)
91°45
′W black spruce
Nipigon Corrigal 49°01
′N 1988 black spruce 1990 BBt, BSg CON, CRg, RCBD: 3 Blocks (18)
88°10
′W EZg, RHg
Nipigon Hele 48°59
′N 1987 black spruce 1990 CON, CRg, RHg, SGh RCBD: 3 Blocks (12)
88°33′W
a
Treatment descriptions: ASg – aerial application of Vision (glyphosate) from a Bell 206 helicopter; ASt – aerial application of Release (triclopyr) from a Bell 206 helicopter; BBt
– basal Bark application of Release (triclopyr) with backpack sprayer (Thin Line); BS – motor-manual brush saw cutting at 18 cm above ground without herbicide; BSg – brush
saw cutting with stump herbicide applicator attachment with Vision (glyphosate); CON – untreated control; CRb – continuous removal of vegetation by annual applications of
brush saws; CRg – continuous removal of vegetation by annual applications of Vision (glyphosate); EZg – EZ-Ject injection of Vision (glyphosate) into competing basal stem;
MBg – Backpack mist blower application of Vision (glyphosate); RHg – Reel and hose application of Vision (glyphosate); SGh – spot gun application of Velpar-L (hexazinone);
SIL – mechanical brush cutting at 33 cm above ground with Silvana Selective/Ford Versatile tractor
1
Hog fuel consists of mix of wood residues, shavings and off-cuts
that can be recovered during saw-log and pulp-log conversion.
The Forestry Chronicle Downloaded from pubs.cif-ifc.org by Lakehead University on 10/28/16
For personal use only.
262
ma r s/av r il
2011, v o l . 87, N
o
2 — Th e For esTr y Ch r oNiCl e
(1986–1988), and planted (1988–1991) with bareroot or con-
tainer stock of jack pine (Pinus banksiana Lamb.) (three sites),
black spruce (Picea mariana [Mill.] BSP) (two sites), or white
spruce (Picea glauca [Moench] Voss) (one site) at approxi-
mately 2 × 2 m spacing. In all, 12 vegetation management
(release) treatments
5
(plus untreated control) were applied to
97 experimental units (plots) varying from 2 ha to 12 ha,
using randomized complete block designs with single replica-
tions. One exception is the Leether Lake site, where treat-
ments were completely randomized in four replications. Each
site has three to four blocks and four to seven treatments, but
all treatments were not applied at all sites. The level of silvi-
culture implemented at all six sites would be classified as
“basic” based on the definitions provided by Bell et al. (2008).
Specific site and treatment descriptions are below.
The Bending Lake Project is located about 54 km north of
Atikokan, Ontario (Table 1). Jack pine is the crop species at
this site. The study includes four blocks and four treatments:
(i) aerial spray with glyphosate (ASg) in late August 1992,
(ii) brush saw (BS) between late June and early July 1993,
(iii) control (CON) with no release treatment, and (iv) con-
tinuous removal with ground applications of glyphosate
(CRg) in September 1993 and again in August 1994.
The Espanola Study is located approximately 90 km north-
west of Espanola, Ontario (Table 1). Jack pine is the crop
species at this site. The study includes three blocks and six treat-
ments: (i) aerial spray with glyphosate (ASg) in August 1993,
(ii) basal bark/triclopyr (BBt) in October 1993, (iii) brush saw
(BS) in October 1993, (iv) control (CON) with no release treat-
ment, (v) continuous removal with ground applications of
glyphosate (CRg) in June 1995 and again in June 1996, and (vi)
mist blower with glyphosate (MBg) in August 1993.
The Fallingsnow Ecosystem Project is located approxi-
mately 60 km southwest of Thunder Bay, Ontario (Table 1).
White spruce is the main crop species at this site. The study
includes three blocks and seven treatments: (i) aerial spray
with glyphosate (ASg) in mid-August 1993, (2) aerial spray
with triclopyr (ASt) in mid-August 1993, (iii) brush saw (BS)
in mid- to late October 1993, (iv) control (CON) with no
release treatment, (v) continuous removal with brush saws
(CRb) in 1994 through 1997, (vi) continuous removal with
ground applications of glyphosate (CRg) in 1994 through
1997, and (vii) Silvana Selective (SIL) brush cutting between
late October to early November 1993.
The Leether Lake Study is located about 56 km north of
Sioux Lookout, Ontario (Table 1). Jack pine is the crop species
at this site except in blocks treated with BS, where black
spruce was planted. The study includes four treatments repli-
cated three times: (i) aerial spray with glyphosate (ASg) in
August 1993, (ii) brush saw (BS) between early to mid-June
1994, (iii) control (CON) with no release treatment, and (iv)
continuous removal with ground applications of glyphosate
(CRg) in 1994 through 1996.
The Nipigon Hele Study is located in Hele Township,
about 19 km west of Nipigon, Ontario (Table 1). Black spruce
is the main crop species at this site. The study includes three
blocks and four treatments: (i) control (CON) with no release
treatments, (ii) continuous removal with ground applications
of glyphosate (CRg) from August 1990 through 1994; (iii) reel
and hose application of glyphosate (RHg) in August 1991; and
(iv) spot gun application of hexazinone (SGh) in October
1990. All treatments except continuous removal were applied
to a 1-m radius around each crop tree.
The Nipigon Corrigal Study is located in Corrigal Town-
ship, about 8 km east of Nipigon, Ontario (Table 1). Black
spruce is the main crop species at this site. The study includes
three blocks and six treatments: (i) basal bark application of
triclopyr (BBt) in October 1990, (ii) brush saw with
glyphosate (BSg) in September 1990, (iii) control (CON) with
no release treatment, (iv) continuous removal with ground
applied glyphosate (CRg) in August 1990 through 1994, (v)
EZ-Ject application of glyphosate (EZg) in November 1990,
and (vi) reel and hose application of glyphosate (RHg) in
August 1991. All treatments except continuous removal, EZ-
Ject, and basal bark were applied to a 1-m radius around each
crop tree.
Data collection
Crop tree plots of approximately 1200 m
2
(30 m × 40 m) were
established in each treatment plot, before applying the release
treatment. In each treatment plot, 20 crop trees (at approxi-
mately 10-m spacing) were selected for periodic remeasure-
ment. We used the 10
th
-year post-treatment crop tree meas-
urement data (height, diameter at breast height (DBH) and
stocking) presented in Bell et al. (2011a) and additional 16
th
-
year post-treatment data from the Fallingsnow Ecosystem
Project collected in 2009 summer for our analyses.
Simulation and optimization models
We used Forest Vegetation Simulator, FVS
Ontario
—a non-spa-
tial, individual-tree growth model (for details, see ESSA
2008)—to project expected crop tree volumes to an arbitrary
rotation age of 70 years. The model simulates changes in
diameter increment of individual trees using current size
(diameter and height) and calibrated values of previous
growth. A sub- model accumulates periodic increments over
successive time intervals (e.g., five or 10 years). For each site,
a common forest region (Ontario West), site quality (Site
quality II), crop species, and establishment year were used.
We simulated total volume assuming equal spacing between
existing trees and no additional silvicultural treatments. The
existing stand condition was defined using the 10
th
-year post-
treatment measurement data for all sites, except Fallingsnow,
for which 16
th
-year post-treatment data were used. Total tree
height, diameter at breast height, number of tree stems per
hectare (SPH), and stocking information were used as inputs
to the simulation model, combining the data from all blocks.
We projected SPH, gross total volume (GTV), gross mer-
chantable volume (GMV), basal area (BA), quadratic mean
diameter (QMD), and top height (TH) of each crop species
for each treatment combination.
BUCK-2 (Zakrzewski et al. 2010) was used to optimize the
possible product mix and estimate the future value of fibre
produced. Projected SPH and GTV, and mean diameter and
top height were used as inputs. In this optimization tool, the
desired size limits (length and minimum diameter) of round-
wood timber products and rankings of log categories (sawlogs,
veneer logs, and pulp wood) are user-defined. Though BUCK-
2 does not account for the price of the output lumber, its objec-
tive is to maximize the total monetary value of a sum of the
5
Vegetation management treatments are sometimes referred as
“release treatments”. These terms are interchangeably used
throughout this paper.
The Forestry Chronicle Downloaded from pubs.cif-ifc.org by Lakehead University on 10/28/16
For personal use only.
ma r Ch /apr il 2011, v ol . 87, No. 2 — Th e Fo r esTr y Ch r oNiCl e 263
user-defined timber products at the tree level, where the proxy
for that value is a user-defined ranking of the product. In other
words, the optimum timber product mix is such that it maxi-
mizes volume proportion of the most profitable timber prod-
uct at the tree level. The constraints are user-defined timber
product sizes (constants): minimum top diameters and fixed
log lengths. We used 2.44 m (8 feet) minimum length and 30
cm minimum diameter for the first category of sawlog (Rank
I), 2.44 m (8 feet) minimum length and 20 cm minimum
diameter for the second category of sawlog (Rank II), 1.22 m
(4 feet) minimum length and 10 cm minimum diameter for
pulp logs (Rank III), and a kerf factor of 1.5 cm (assuming
wastage allowance for circular saw), as constraints to optimize
the proportion of wood products expected from the projected
Table 2. Treatment costs, projected gross total volume and merchantable volume of crop tree species at 70 years by study site
and treatment.
Study sites
Bending Leether Nipigon Nipigon
Metric Treatment
a
Lake Espanola Fallingsnow Lake Hele Corrigal Average
Costs ASg 202.0 240.0 190.0 210.0 ––210.5
(CAD$ ha
-1
)
b
ASt ––268.9 –––268.9
BBt – 535.0 –––455.0 495.0
BS 550.0 600.0 500.0 550.0 ––550.3
BSg –––––625.3 625.3
CON 0.0 0.0 0.0 0.0 0.0 0.0 0.0
CRb ––1750.0 –––1750.0
CRg 1124.8 1162.8 2011.8 1132.8 2248.9 2248.9 1655.0
EZg –––––910.3 910.3
MBg – 475.3 ––––475.3
RHg ––––426.6 395.0 410.8
SGh ––––675.3 – 675.3
SIL ––650.3 –––650.3
GTV ASg 205 232 260 208 ––226.3
(m
3
ha
-1
)
c
ASt ––287 –––287.5
BBt – 165 –––169 167.0
BS 167 172 248 169 ––189.0
BSg –––––177 177.0
CON 182 181 204 179 131 153 171.6
CRb ––299 –––299.0
CRg 288 280 302 296 180 190 256.1
EZg –––––170 170.3
MBg – 251 ––––250.7
RHg ––––196 186 191.1
SGh ––––169 – 168.7
SIL ––255 –––255.0
Merchantable volume ASg 173 (5) 186 (5) 208 (51) 185 (157) ––188 (54)
(m
3
ha
-1
)
c,d
BBt – 132 (70) –––132 (147) 132 (108)
BS 159 (3) 138 (6) 199 (114) 142 (72) ––159 (49)
BSg –––––138 (120) 138 (120)
CON 154 (60) 145 (109) 163 (172) 153 (252) 104 (60) 124 (100) 140 (126)
CRb ––260 (9) –––260 (9)
CRg 244 (2) 224 (2) 266 (1) 254 (73) 148 (2) 150 (16) 214 (16)
EZg –––––138 (218) 138 (218)
MBg – 201 (11) ––––201 (11)
RHg ––––155 (23) 147 (53) 151 (38)
SGh ––––133 (91) – 133 (91)
SIL ––204 (135) –––204 (135)
a
Treatment descriptions: ASg – aerial application of Vision (glyphosate) from a Bell 206 helicopter; ASt – aerial application of Release (triclopyr) from a Bell 206 helicopter; BBt –
basal Bark application of Release (triclopyr) with backpack sprayer (Thin Line); BS – motor-manual brush saw cutting at 18 cm above ground without herbicide; BSg – brush saw
cutting with stump herbicide applicator attachment with Vision (glyphosate); CON – untreated control; CRb – continuous removal of vegetation by annual applications of brush
saws; CRg – continuous removal of vegetation by annual applications of Vision (glyphosate); EZg – EZ-Ject injection of Vision (glyphosate) into competing basal stem; MBg – Back-
pack mist blower application of Vision (glyphosate); RHg – Reel and hose application of Vision (glyphosate); SGh – spot gun application of Velpar-L (hexazinone); SIL – mechanical
brush cutting at 33 cm above ground with Silvana Selective/Ford Versatile tractor
b
Costs are based on non-crop stocking level and distance of the site from the nearest major centre and are calculated on a 500-ha plot basis.
c
Volumes were projected using FVS
Ontario
.
d
Values in parentheses are the non-crop (hardwoods – mainly poplar) merchantable volumes.
The Forestry Chronicle Downloaded from pubs.cif-ifc.org by Lakehead University on 10/28/16
For personal use only.