Michigan Technological University
Digital Commons @ Michigan Tech
Dissertations, Master's Theses and Master's
Reports - Open
Dissertations, Master's Theses and Master's
Reports
2011
Variation in carbon content of tropical tree species from Ghana
Daniel Yeboah
Michigan Technological University
Follow this and additional works at: https://digitalcommons.mtu.edu/etds
Part of the Forest Sciences Commons
Copyright 2011 Daniel Yeboah
Recommended Citation
Yeboah, Daniel, "Variation in carbon content of tropical tree species from Ghana", Master's Thesis,
Michigan Technological University, 2011.
https://doi.org/10.37099/mtu.dc.etds/161
Follow this and additional works at: https://digitalcommons.mtu.edu/etds
Part of the Forest Sciences Commons
VARIATION IN CARBON CONTENT OF TROPICAL TREE SPECIES FROM
GHANA
By
Daniel Yeboah
A THESIS
Submitted in partial fulfillment of the requirements for the degree of
MASTER OF SCIENCE
(Applied Ecology)
MICHIGAN TECHNOLOGICAL UNIVERSITY
2011
© 2011 Daniel Yeboah
This thesis, “Variation in Carbon Content of Tropical Tree Species from Ghana,’’ is
hereby approved in partial fulfillment of the requirements for the Degree of MASTER
OF SCIENCE IN APPLIED ECOLOGY.
School of Forest Resources and Environmental Science
Signatures:
Thesis Co-advisor ______________________________________
Dr. Andrew J. Burton
Thesis Co-advisor ______________________________________
Dr. Andrew J. Storer
Dean _______________________________________
Dr. Margaret R. Gale
Date _______________________________________
Table of Contents
List of Figures ..................................................................................................................... 4
List of Tables ...................................................................................................................... 6
Acknowledgements ............................................................................................................. 7
Thesis Abstract.................................................................................................................... 8
Chapter 1 Introduction ........................................................................................................ 9
Chapter 2 Variation in carbon content of tropical tree species from Ghana ..................... 16
Abstract ................................................................................................................. 16
Introduction ........................................................................................................... 17
Methods................................................................................................................. 21
Results ................................................................................................................... 25
Discussion ............................................................................................................. 49
Conclusion and Recommendations ....................................................................... 55
Literature Cited ................................................................................................................. 58
3
List of Figures
Figure 2.1: Carbon concentration of 12 year-old-tree species from the OCAP
plantation in the wet evergreen forest ecozone of Ghana ......................................31
Figure 2.2: Comparison of C concentration for 7 and 12 year-old trees of the
same species from OCAP plantations in the wet evergreen forest ecozone
of Ghana .................................................................................................................32
Figure 2.3: Mean volume of species in a 12 year-old plantation at OCAP in the
wet evergreen forest ecozone of Ghana .................................................................33
Figure 2.4: Wood density estimate of trees species from a 12 year-old plantation
at OCAP in the wet evergreen forest ecozone of Ghana .......................................34
Figure 2.5: Comparison of wood density in trees from 12 and 7 year-old
plantations at OCAP in the wet evergreen forest ecozone in Ghana .....................35
Figure 2.6: Comparison of wood density in 7 year-old Khaya spp from a location
at OCAP in the wet evergreen forest ecozone and Bobiri in the moist
semi-deciduous ecozone of Ghana.........................................................................36
Figure 2.7: Mean C content per tree for species growing in a 12 year-old
plantation at OCAP in the wet evergreen forest ecozone of Ghana ......................37
Figure 2.8: Allometric relationship between volume and dbh (power function)
for 18 trees species from 12 year-old plantation at OCAP in wet
evergreen forest ecozone of Ghana .......................................................................41
4
Figure 2.9: Ln-ln relationship between dbh and volume for 18 trees species from a
12-year-old plantation at OCAP in the wet evergreen forest ecozone of Ghana ...42
Figure 2.10: Allometric relationship (power function) between volume and
dbh2×height (D2H) for 18 trees species from a 12 year-old plantation at OCAP in
the wet evergreen forest ecozone of Ghana ..........................................................43
Figure 2.11: Ln-ln relationship between volume and dbh2×height (D2H) for 18 trees
species from a 12-year-old plantation at OCAP in the wet evergreen forest
ecozone of Ghana ...................................................................................................44
Figure 2.12: Allometric relationship between carbon content and dbh for 18 trees species
from 12 year-old plantation at OCAP in the wet evergreen forest ecozone of
Ghana .....................................................................................................................45
Figure 2.13: Allometric relationships between trees of high volume and dbh, and low
volume and dbh (power function) from OCAP plantation in the wet evergreen
forest ecozone of Ghana.........................................................................................46
Figure 2.14: Allometric relationships between trees of high biomass and dbh, and low
biomass and dbh (power function) from OCAP plantation in the wet evergreen
forest ecozone of Ghana.........................................................................................47
Figure 2.15: Allometric relationship between trees of high C content and dbh, and low C
content and dbh (power function) from OCAP plantation in the wet evergreen
forest ecozone of Ghana.........................................................................................48
5
List of Tables
Table 2.1: Mean C concentration for 12-year- old trees from OCAP plantation in the wet
evergreen forest ecozone of Ghana ........................................................................26
Table 2.2: Mean wood density for 12-year-old trees from OCAP plantation in the wet
forest ecozone of Ghana and densities in literature ...............................................28
Table 2.3: Summary of C sequestration for tree species from OCAP plantation in the wet
evergreen and Bobiri in the moist semi-deciduous forest ecozones of Ghana ......38
Table 2.4: Regression equations for volume (m3) and biomass (kg) for trees from OCAP
plantation in the wet evergreen forest ecozone of Ghana ......................................40
6
Acknowledgements
My utmost thanks go to my advisors, Dr. Andrew J. Storer and Dr. Andrew J. Burton for
their advice, support and teaching that has brought me up to the level of a scientist and a
professional. I thank my committee member, Dr. Amy M. Marcarelli for her advice and
help throughout my research project. My profound gratitude to Dr. Emmanuel OpuniFrimpong, Forestry Research Institute of Ghana, for serving as local advisor in Ghana
and his team of students who assisted with data collection. I am grateful to the late Dr.
David F. Karnosky, for his contribution in the development of this research concept. I
also thank the Dean of the School of Forest Resources and Environmental Science, Dr.
Margaret R. Gale for her efforts in recruiting Ghanaian students; your encouragement and
dedicated professors made learning in the school enjoyable. A million thanks go to all the
field technicians and managers of my former employers in Ghana, Samartex Timber and
Plywood Ltd that assisted with data collection. My sincere gratitude goes to my
colleague, Emmanuel Ebanyenle, for sharing key information that helped in my research.
I also thank everyone who has contributed in diverse ways to this research project that
helped create a wonderful and enjoyable research experience for everyone involved. I
appreciate my wife, Millicent Yeboah and children for their support. Above all, I thank
God for all the blessings and opportunity to be part of this great family at Michigan Tech.
7
Abstract
Most research on carbon content of trees has focused on temperate tree species with little
information existing on the carbon content of tropical tree species. This study
investigated the variation in carbon content of selected tropical tree species and compared
carbon content of Khaya spp from two ecozones in Ghana. Allometric equations
developed for mixed-plantation stands for wet evergreen forest verified the expected
strong relationship between tree volumes and dbh (r2>0.93) and volume and dbh2×height
(r2>0.97). Carbon concentration, wood density and carbon content differed significantly
among species. Volume at age 12 ranged from 0.01 to 1.04 m3 per tree, and wood density
was highly variable among species, ranging from 0.27 to 0.76 g cm-3. This suggests that
species specific density data is critical for accurate conversion of volumes derived from
allometric relationships into carbon contents. Significant differences in density of Khaya
spp existed between the wet and moist semi-deciduous ecozones. The baseline specieslevel information from this study will be useful for carbon accounting and development
of carbon sequestration strategies in Ghana and other tropical African countries.
8
Chapter 1
Introduction
Forest Estate of Ghana
Ghana is endowed with an extensive stretch of tropical forests characterized by diverse
flora and fauna (Abebrese 2002). The total forest cover is about 8.2 million ha and
consists of two forest types: forest reserve and off-reserve forest (Hall and Swaine 1976;
Hall and Swaine 1981). These two forest types differ in many ways. The off-reserve
forest is on land primarily used for agricultural purposes, and accounts for about 6.5
million ha, or about 70% of the total forests, which cover 8.2 million ha (Hall and Swaine
1976; Hall and Swaine 1981; Abebrese 2002). The reserve forests constitute about 1.2
million ha and are dedicated only for forestry purposes (Abebrese 2002).The reserve
forest is further classified into wet evergreen, moist evergreen, upland evergreen, moist
semi-deciduous, dry semi-deciduous, southern marginal, and south-east outlier,
depending on rainfall regime (Hall and Swaine 1976). The diversity of species associated
with each of these forest types is high, and each contains unique species assemblages. For
example the moist-evergreen forest contains about 250 tree species per ha (Hall and
Swaine 1981). The reserve and off-reserve forests serve as an economic, ecological and
environmental asset to Ghana. The forestry sector employs about 75,000 people and
contributes 6% to 8% to the country’s Gross Domestic Product (Atuahene 2001).
9
Unsustainable anthropogenic activities that are known to devastate the off-reserve forest
include unsustainable logging, uncontrolled fire, and conversion of forests to agricultural
lands (Hawthorne and Abu-Juam 1995). International demand for timber after the Second
World War also led to an expansion of timber industries in Ghana which intensified the
depletion of the forest resources especially in the off-reserve forest (Hawthorne and AbuJuam, 1995; Amanor 1997; Nanang 2010). Between 1970 and 1990, Ghana lost 1.3% of
its forest each year as a result of harvesting and degradation (Dixon et al. 1996). The offreserve forest has therefore been depleted due to a growing population and because forest
products and services are of low value compared to non-forest products produced after
converting the forest land. For example, people often convert forestlands to cocoa
plantations, which they perceive to be of immediate benefit (Hall and Swaine 1981;
Hawthorne and Abu-Juam 1995; Amanor 1997).Despite the loss of the off-reserve forest,
the reserve forest remains largely intact (Amanor 1997; Sandker et al. 2010).
Atmospheric CO2 and Climate Change
During pre-industrial times, atmospheric CO2 concentration was about 280ppm but this
has increased substantially to about 368ppm by 2000 (Malhi et al. 2002). This is largely
attributed to emissions from burning of fossil fuel and vegetation (Malhi et al. 2002).
This elevated atmospheric CO2 is a contributor to global climate change, which has
increased the average global temperature by about 0.74 oC over the past hundred years
(IPCC 2007).
Tropical forests have been recognized for their potential to store carbon in biomass and
help ameliorate the rising level of CO2. Brown and Lugo (1982) noted that tropical
10
forests could be credited with about 20% of the total carbon budget of the world. The
forests of Ghana contain biomass of about 1,132 MtC (FAO 2005) and the forests of
Africa contain 60 GtC of biomass (FAO 2005). World soil organic matter harbors 1,500
to 2,100 Pg carbon and terrestrial plants contain 490 to 760 Pg carbon, compared to 760
Pg carbon in the atmosphere (Amthor 1995).
Many initiatives and efforts seek to harness the carbon storage capability of tropical
forests. The Kyoto Protocol is an international initiative geared towards finding solutions
to concerns of global warming and was adopted in 1997 by member states. To date, 193
parties, made up of 192 countries and 1 regional economic integration organisation have
ratified the Kyoto Protocol (UNFCCC 1997). The Protocol placed emphases on
commitments of member states to reduce their CO2 emission by sequestering carbon in
forestry and agriculture systems through Clean Development Mechanisms (UNFCCC
1997 ). It is therefore implied that a developed nation could partner and sponsor
reforestation or afforestation projects in developing countries and obtain carbon credits
(UNFCCC 1997). Major CO2 producing countries were initially unwilling to ratify the
Kyoto Protocol which delayed the commencement of the Protocol because the minimum
number of member countries required for achieving a target of at least 55% reduction of
CO2 emissions had not been satisfied (UNFCCC 1997). However, the European Union
and Japan signed and were followed by Canada in 2002. The threshold for the Kyoto
Protocol to become binding was reached in 2005, when Russia, which accounted for 17%
of the world’s CO2 emission in 2004, ratified the Protocol (UNFCCC 1997 ). Ghana has
also signed the Kyoto Protocol and is committed to fulfil the obligations documented in
the Protocol.
11
Management and Carbon sequestration
Forest management activities have serious repercussions for forest carbon stocks. For
example, irrigation, thinning, and fertilizer application are key management actions,
which can boost forest productivity and carbon stocks. Mean carbon stocks may increase
following cumulative nitrogen fertilization (Hyvonen et al. 2008). Fire occurrence in
forests, soil compaction during tillage and animal grazing also influences forest carbon
stocks. Fire increases forest floor debris and releases soil carbon, and has been reported to
increase coarse-woody debris in young forests compared to mature stands (Litton et al.
2004). This destruction can increase both heterotrophic respiration and ecosystem carbon
loss (Barnes et al. 1998).
Forestry activities that increase stand density may increase below and above ground
carbon (Litton et al. 2004). In addition, prolonged stand rotations result in higher carbon
sequestration than do shorter rotations (Schroeder 1992).
Expansion of forests by embarking on reforestation and afforestation projects holds great
potential for storing carbon in biomass in tropical regions (Winjum and Schroeder 1997;
Nair et al. 2009). Restocking of degraded forests through enrichment planting programs
and agroforestry intervention enhances carbon storage of forests (Schroeder 1992; Nair et
al. 2009).
Carbon Projects in Africa
Many afforestation and reforestation projects have been executed in Africa as a means to
sequester CO2 in biomass and provide carbon credits for participants. Carbon trading
provides an attractive economic opportunity for subsistence farmers to sell sequestered
12
carbon to interested partners in industrialized nations. An initiative by the World Bank
has funded twelve projects in Africa through its BioCarbon fund and Global Environment
facility (Jindal et al. 2008). For example, the World Bank funded a Nile Basin
reforestation project in Uganda, where about 2000ha in plantations were established with
timber and carbon credits shared between local communities and the bank (Jindal et al.
2008). A similar project was funded by the United States Agency for International
Development (USAID). The European Union and FACE foundation are also funding
other carbon projects (Jindal et al. 2008).
Carbon Analysis
Carbon (C) concentration of dry wood has generally been assumed to be 50% for most
species (Matthews 1993). However, wood is comprised of a wide range of
macromolecular substances such as lignin, cellulose and hemicellulose. There are varying
proportions of C in each of these compounds and compound groups (Lamlom and
Savidge 2006). Lamlom and Savidge (2006) reported that there is 42.1% carbon in
cellobiose, the building blocks of cellulose, and 40% C in monosaccharides that are
associated with hemicellulose. Different plant tissues contain varying amounts of carbon.
For example, carbon contents of leaves is about, 42%, while roots contain 47 to 52%
carbon (Atjay et al. 1979; Lamlom and Savidge 2006).
Tree Biomass and Allometry
Methods for estimating tree biomass have attracted much scientific attention recently
because of their importance in estimating forests carbon stocks (Zianis and Mencuccini
13
2004). Biomass can be calculated from knowledge of both the volume and density of a
tree (Zobel and van Buijtenen 1989; Brown 1997; Ketterings et al. 2001).
Using volume alone to estimate biomass may not accurately estimate the amount of
substance per unit area, as it ignores significant variation in density among species (Zobel
and van Buijtenen 1989; Brown 1997). Organic carbon occurs in various pools within the
forest ecosystem: above and below ground biomass, woody debris, mineral soils, forest
floor and heterotrophic organisms (Barnes et al. 1998). A study in the United Kingdom
showed that a plantation may hold carbon in the range of about 40-80 Mg Cha-1 in trees,
15-25 Mg Cha-1 in above and belowground litter, and 70-90 Mg Cha-1 in soil organic
matter (Dewar and Cannell 1992). Above ground biomass is usually estimated with a
widely applied power function model of the form: M = aDb, where, a and b represent
scaling coefficients, D is the diameter at breast height and M is total aboveground tree
dry biomass (Ketterings et al. 2001; Zianis and Mencuccini 2004). The values of the two
scaling coefficients vary with species, stand age, site quality, and climate and stand
stocking (Baskerville 1965; Zianis and Mencuccini 2004). To develop allometric
equations, trees are cut down from a forest stand to measure diameter at breast height
(dbh) and height, which are used to estimate volumes. The calculated volumes are
regressed on either dbh or the combination of dbh and height to establish the allometric
equation (Brown et al. 1989; Ketterings et al. 2001; Zianis and Mencuccini 2004). The
developed equation could be applied to estimate volume of all trees within an entire area
based on either dbh or dbh and height, which can then be converted to biomass using
wood density.
14
In this thesis, the biomass and carbon content of 18 tree species from tropical forest
plantations in Ghana were estimated from wood samples collected in wet and moist forest
ecozones. Species-specific information on carbon concentration, and wood density and
methods for calculating carbon content are described, and these will be useful for both
commercial forest plantations and reforestation activities.
15
Chapter 2
Variation in carbon content of tropical tree species from Ghana
Abstract
Most research on the carbon content of trees has focused on temperate tree species with
little information existing for tropical tree species. Questions remain regarding how much
carbon can be sequestered by various tree species and in different forest climatic zones.
This study was designed to investigate the variation in carbon content of selected tropical
tree species and compare the carbon content of Khaya spp from two ecozones in Ghana.
Two to three individuals of 18 tree species were randomly selected and harvested from
12-year-old and 7-year-old plantations in Ghana. The diameter at breast height (dbh at
1.3 m above ground) and length of the main stem were measured. A 2-cm thick wood
disc was cut from the bottom, middle and top positions of the main stem of each tree, and
used to estimate wood density and carbon concentration. Estimates of tree stem carbon
were computed using tree stem volumes, density and carbon concentration. Allometric
equations developed for mixed plantation stands for the wet evergreen forest verified the
expected strong relationship between tree stem volumes and dbh (r2>0.93) and between
volumes and dbh2×height (r2>0.97). Carbon concentration, wood density and carbon
content differed significantly among tree species. Carbon concentration of the tree
species ranged from 46.3 to 48.9 %. Volume for the 12-year-old trees varied widely
among species, from 0.01 m3 to 1.04 m3. Wood densities differed among tree species and
16
the three stem positions. Differences in wood density at the three positions on the stem
were independent of tree species. Wood density was highly variable among species,
ranging from 0.27 g cm-3 to 0.76 g cm-3. Species specific knowledge of wood density was
much more important than knowledge of carbon concentration for ensuring accurate
conversion of allometric volume estimates to tree carbon content. Significant differences
in wood density did exist among Khaya spp from wet and moist semi-deciduous
ecozones, suggesting climatic factors may also need to be considered. This study has
provided baseline species-level information that will be useful for carbon accounting and
development of carbon sequestration strategies in Ghana and other tropical African
countries.
Introduction
Growing concerns about climate change resulting from increased concentration of
greenhouse gases in the atmosphere have stimulated discussions about the importance
and potential of forests for carbon sequestration. Due to anthropogenic emissions, the
concentration of the major greenhouse gas, carbon dioxide (CO2), has increased from 290
to 390 ppm within the last hundred years (Schneider 1990). Mean global temperatures
have increased by 0.74 oC over the same time period, as atmospheric CO2 concentration
increased (IPCC 2007). Regional temperatures may increase even by 1 to 5 ºC, if the
current atmospheric CO2 concentration is doubled (Mahlman 1997).
To reduce the escalating levels of greenhouse gases, in particular CO2, afforestation and
reforestation systems have been encouraged as means to sequester CO2 in biomass, an
17
idea formally endorsed by the Kyoto Protocol. The Kyoto Protocol allows for the
opportunity to offset CO2 emissions through collaboration between developed and
developing nations to venture into reforestation or afforestation projects (UNFCCC
1997).
Questions regarding how much carbon can be sequestered by different tree species and if
there are variations in carbon content of trees with geographical location remain to be
answered. Available research has revealed significant differences among different tree
species growing at various sites (Elias and Potvin 2003; Lamlom and Savidge 2003; Bert
and Danjon 2006). The chemical make-up of different tree species allows them to grow
in different environments (Elias and Potvin 2003; Lamlom and Savidge 2003; Bert and
Danjon 2006) and results in variation in carbon content between species and at different
locations for trees of similar size.
Most research estimating carbon content of trees has focused on temperate trees. Little
information on the carbon content for tropical trees species exists, and such paucity of
information makes estimation of the value of these species as carbon sinks difficult.
Quantifying carbon stocks in forests requires accurate estimation of aboveground
biomass in addition to information about the carbon concentration (Brown et al. 1989;
Ketterings et al. 2001; Elias and Potvin 2003; Lamlom and Savidge 2003; Chave et al.
2004). Several factors account for variability in tree and forest biomass, including tree
species, climate, topography, soil fertility, water supply, and wood density (Fearnside
1997; Luizao et al. 2004; Sicard et al. 2006; Slik et al. 2008). Wood density is an
important variable which affects biomass estimates derived by converting volumes from
forest inventory data (Brown et al. 1989; Fearnside 1997). Tree species mass is known to
18
be influenced by factors such as architecture, size, form, health, and variation of wood
density (Basuki et al. 2009). Along the main stem of a tree, wood density varies from the
base to the top of the stem, and radially from pith to bark. Wood density often decreases
from the stump to half of the total height of the tree, and increases afterwards towards the
top (Espinoza 2004). Density also varies with species, age and geographical location in
tropical forests (Fearnside 1997; Slik et al. 2008; Henry et al. 2010). However, little
information exists on wood density for plantation tree species grown in Ghana and other
sub-Saharan African countries.
Forest biomass estimation usually involves conducting forest inventory on sampled plots,
using appropriate allometric equations to estimate tree volumes, converting volumes to
biomass using wood density, and extrapolating to estimate biomass for an entire area
(Brown 1997; Ketterings et al. 2001; Chave et al. 2004). The allometric equation used is
the most essential input from this method (Navar 2009). Development of these equations
is achieved by fitted equations using regression techniques (Parresol 1999; Wirth et al.
2004). There is a possibility of error in above ground biomass estimation by inappropriate
application of the same allometric equation to different forests types (Brown et al. 1989;
Clark and Clark 2000). For example, the equation may be developed based on a limited
size class of trees which skews the equation towards this size class (Clark et al. 2001;
Henry et al. 2010) and could introduce error when applied to trees that fall outside the
range of sizes used to develop the equation. Literature is replete with several allometric
equations for estimating aboveground biomass of some tropical forests (Brown et al.
1989; Chave et al. 2005). In Ghana, however, allometric equations rarely exist, especially
19
for plantation grown trees. Hence, the option is to apply equations from other regions, the
reliability of which has not been tested for Ghana (Brown et al. 1989; Henry et al. 2010).
This research seeks to bridge the knowledge gap on the carbon content of tropical trees
species from Ghana by developing allometric equations applicable for plantation grown
tree species and providing information on carbon concentration, wood density, and tree
carbon content. The study investigated eighteen fast growing species common to the
moist semi-deciduous and wet forest ecozones of Ghana. The purpose of promoting
plantation development in Ghana is to restore degraded forests, provide raw materials for
industry and potentially obtain extra income from carbon credits as a means of value
addition. The following research questions were addressed:
•
What is the estimated average carbon content in stems of selected
plantation trees species grown in Ghana?
•
What is the variation in carbon content among the different tree species
grown in Ghana, and how is this affected by species differences in
volume, density and carbon concentration?
•
What is the variation in carbon content within a species from two different
ecological zones in Ghana?
Hypothesis
1. There are significant differences in carbon content among different tree
species due primarily to differences in wood density.
20
2. There are significant differences in carbon content of the same trees
species planted in different ecological zones with greater carbon content
occurring in wetter zones due to differences in wood density.
Objectives
1. To estimate carbon content in selected tropical trees in Ghana.
2. To compare carbon content of plantation trees from moist semi-deciduous
and wet evergreen forest zones of Ghana, and determine which species
have the greatest carbon sequestration potential.
Methods
Study Area
Two study areas were used, the first in Oda-kotoamso and the second at Bobiri forest
reserve. Oda kotoamso is located in the western region of Ghana, and is about 10 km
from Asankraqwa, the district capital of Wassa Amenfi.
Geographically, Odokoamso lies between latitude 5º 18’N and 5º 45’N and longitude
2º10’W and 2º30’W. Oda kotoamso falls within the hot humid tropical rainforest of the
wet evergreen forest zone of Ghana (Hall and Swaine 1981). There are two rainfall
seasons: a major rainy season from April to July, and a minor season from August to
September. Average annual rainfall ranges from 1750 to 2000 mm (Hall and Swaine
1981). Two dry seasons prevail in the area: a major dryseason from December to March,
and a minor dry season from October to November. The soil is acidic with pH of about 3
21
to 4 (Hall and Swaine 1981). Average annual temperature range between 28 and 32 ºC
and relative humidity is about 70% to85%. The landscape of the area is characterised by
undulating stretches of land with hilly and flattened mountains with an elevation ranging
from about 90 to 400 m above sea level.
The plantation called Oda-kotoamso Community Agroforestry Project (OCAP) was
planted in 1997 and has a total size of approximately 290 ha. To date, 23 tropical and
exotic species have been successfully planted as either mixed or single species stands,
with spacing from 3×3 to 4×4 m. The plantation was developed and is owned by over
eighty outgrower farmers with technical and financial support from Samartex Timber and
Plywood Company (Samreboi, Ghana).
The second site for the study was Bobiri (6º40’N, 1º19’W), about 35 km from Kumasi in
the Ashanti region of Ghana. Bobiri falls within the moist semi-deciduous forest, which
is drier than the wet evergreen forest zone (Hall and Swaine 1981). Average annual
rainfall for the moist evergreen forest ranges from 1200 to1800 mm (Hall and Swaine
1981) and the temperature is about 32 ºC. Topography of the area is moderately high with
an elevation of about 150 to 600 m (Hall and Swaine 1981). The soil is slightly acidic
with pH of about 5 to 6 (Hall and Swaine 1981). The Bobiri plantation consists of single
species stand of Khaya ivorensis and Khaya grandifoliola on a one-hectare plot.
Data Collection
A total of sixty-six trees were randomly selected from the plantations in the wet
evergreen and moist semi-deciduous forest zones. For OCAP, trees of 41, 9 and 8 from
12, 7 and 5 years-old were selected from the plantations respectively. In all cases, 2 to 3
trees per species were examined within an age class at a plantation.
22
Tree species collected for study from the wet evergreen forest zone at OCAP were:
Aningeria robusta, Pycnanthus angolense, Tectona grandis, Cedrela odorata, Heritiera
utilis, Antiaris toxicaria, Tieghemelia heckelii, Ceiba pentandra, Terminalia ivorensis,
Terminalia superba, Milicia excels, Lophira elata, Triplochiton scleroxylem, Mammea
Africana, Guarea thompsonii, Khaya ivorensis, khaya gramdifoliola, Turreanthus
africanus. Eight trees of Khaya ivorensis and Khaya grandifoliola were also selected
from the moist semi-deciduous forest zone at Bobiri.
The trees were cut down and their diameter at breast height (dbh, at 1.3 m) was
measured. The length of the main stem from bottom to top (stump to first large branch) of
individual trees was measured. Volumes of the base (stump to 1.3 m), middle (1.3 m to
midpoint) and top segments (midpoint to top) were computed using Smalian’s formula
(Avery and Burkhart 2002) using the length and end diameters of each segment. Discs of
about 2 cm thickness were cut at the base, middle, and top portion of the main stem of
each tree and their diameter outside bark was measured. Strip sections of wood along the
diameter of the discs were removed as samples. Volumes of these sub-samples were
measured by a water displacement method. This method involves fixing the removed
wedged-shaped samples on a prong attached to an adjustable clamp and submerging them
into a bowl of water placed on an electronic balance. The suspended wood sample in the
bowl of water was fully covered but not touching the bottom of the bowl. Sample volume
(cm3) was determined as the increase in balance reading (g) due to the suspended wood
sample. The wood samples were then kept in airtight bags and stored in a freezer (0 ºC) at
the Forest Research Institute of Ghana, until samples could be transported to Michigan
23
Technological University. Wood samples were oven dried at 70 ºC and weighed with an
electronic scale to determine dry sample weights.
The dry weight of the wood and volume were used to determine density (g cm-3), and the
samples were ground to a fine powder using a ball mill (Spex certi-Prep 8000M).
Samples of the ground wood were then analyzed for carbon concentration using an
elemental analyzer (Fisons NA 1500). The procedure used for estimating the carbon
content of wood was slightly modified from similar work done by Lamlon and Savidge
(2003). Density,volume and carbon concentration were used to estimate carbon content
by segment, with values for the three segments summed to estimate carbon content of the
entire main stem.
Data Analysis
Analysis of variance (ANOVA) was used to test for differences in carbon contents, C
concentration and density of tree species investigated. Two-way ANOVA was used to
test for effects of stem positions, species and their interactions in the analyses of C
concentration and wood density. These analyses were performed in SAS (1997).
Contrasts among tree species were performed using Tukey’s pair-wise comparison for
equal sample sizes, while, Bonferroni’s test was used for unequal sample size at P<0.05.
The main stem height and dbh data were collected were used to develop regression
relationships for predicting volume from dbh and dbh2 ×height. Both linear forms and
power functions were used. The power functions were in the form: M= a(X) b, where, M=
volume of trees, X is either the diameter at breast (dbh) or dbh2×height, and a and b are
scaling coefficients (Zianis and Mencuccini, 2004).
24
Results
Carbon Concentration
Carbon concentration varied significantly among species for both the 12-year-old
(P<0.001) and 7-year-old (P<0.001) plantations. Mean carbon concentrations ranged
from 46.3% to 48.9% (Figure 2.1; Table 2.1). Carbon concentrations were higher for 12year-old trees than 7-year-old trees, but the difference was not statistically significant
(Figure 2.2). Differences in the carbon concentration for the bottom, middle and top stem
positions of 12-year-old trees were significant (P<0.001), but no interactions existed
between species and the stem positions. Regression analysis showed a strong relationship
(P<0.0008; r=0.5020) between carbon concentration and wood density for the 12-yearold trees from the wet evergreen forests ecozone.
25
Table 2.1
Mean carbon concentration for 12-year-r trees from OCAP
plantation the wet evergreen forest ecozone of Ghana
Tree
species
Ceiba pentandra
Heritiera utilis
Tectona grandis
Entandrophragma
angolense
Terminalia ivorensis
Terminalia superba
Milicia excelsa
Mammea africana
Khaya ivorensis
Pycnanthus angolense
Cedrela odorata
Guarea thompsonii
Lophira elata
Turreanthus africanus
Aningeria robusta
Antiaris toxicaria
Tieghemelia heckelii
Triplochiton scleroxylem
Mean C
concentration
(%)
46.8
48.5
48.9
Standard
error of
mean(SE)
0.3
0.3
0.4
46.3
48.0
46.7
46.6
48.6
47.2
46.2
48.3
47.3
48.4
48.6
48.0
47.4
47.7
47.2
0.4
0.2
0.4
0.3
0.1
0.2
0.2
0.2
0.1
0.1
0.1
0.3
0.2
0.3
0.4
Volume and Density
Average tree volume for 12 year-old plantation tree species ranged from a minimum of
0.01 m3 to a maximum of 1.04 m3 (Figure 2.3). The species with the greatest volume at
age 12 was Ceiba pentandra, while Guarea thompsonii had the lowest volume. Also,
wood density differed significantly among species in the 12 year-old plantation (P<0.001;
Figure 2.4). The mean density was highly variable among species, ranging from 0.26 to
0.76 g cm-3(Table 2.2). Ceiba pentandra had the lowest density, while Lophira elata had
the highest density (Figure 2.4). Comparison of wood density of the same species from
12-year-old and 7-year-old plantation in the wet ecozone found not significant
26
differences among the two ages (Figure 2.5). There was a significant difference in wood
density of Khaya spp of the same age but planted in different ecozones (Figure 2.6).
Density also differed significantly along the bole of the trees. Generally, the bottom
positions of trees had higher mean wood density of 0.526±0.02 (SE) g cm-3 than the
middle and the top, which had mean densities of 0.444±0.02 (SE) g cm-3 and 0.439±0.02
(SE) g cm-3 respectively. However, no significant interactions were found between
species and the three stem positions tested with two-way ANOVA. Volume of 12-yearold trees planted in the wet forest ecozone were negatively correlated (P<0.003; r = 0.5403) to wood density.
Wood density inversely correlated with either tree’s main stem height (P<0.0001; r= 0.6279) and dbh (P<0.0006; r= -0.5114) for the 12- year-old trees from the wet evergreen
forest ecozone.
27
Table 2.2
Mean wood density for 12 years trees from OCAP plantation in the wet evergreen forest
ecozone of Ghana and densities in literature.
Wood
density
Tree
in this study
Species
Aningeria robusta
Antiaris toxicaria
Cedrela odorata
(g cm-3)
0.497±0.03
0.356±0.02
0.381±0.03
Ceiba pentandra
Entandrophragma
angolense
Guarea thompsonii
Heritiera utilis
Khaya ivorensis
Lophira elata
Mammea africana
Milicia excelsa
Pycnanthus angolensis
Tectona grandis
Terminalia ivorensis
Terminalia superba
Tieghemella heckelii
Triplochiton scleroxylon
Turreanthus africanus
0.273±0.01
0.439±0.04
Wood density in literature
Reyes et al.
Bolza and Keating
1992
1972
0.526±0.03
0.464±0.03
0.523±0.03
0.761±0.03
0.622±0.01
0.458±0.05
0.354±0.03
0.566±0.02
0.381±0.02
0.419±0.03
0.581±0.05
0.429±0.05
0.435±0.01
28
(g cm-3)
0.380
0.430,0.440
0.450
0.260
(g cm-3)
0.370 to 0.400
0.370 to 0.400
0.270 to 0.320
0.450
0.550
0.560
0.440
0.870
0.620
0.400
0.500,0.550+
0.450
0.550
0.320
-
0.510 to 0.570
0.580 to 0.640
0.580 to 0.640
0.460 to 0.500
0.102 to 0.114
0.650 to 0.720
0.410 to 0.450
0.580 to 0.640
0.510 to 0.570
0.410 to 0.450
0.580 to 0.640
0.370 to 0.400
0.460 to 0.500
Carbon content and Guilds Classification
The mean carbon content across species at age 12 was 54.89±8.16 SE kg C/tree. Guarea
thompsoni had an extremely low average carbon content of 2kg C/tree whereas Ceiba
pentandra sequestered the greatest carbon of 179 kg C/tree (Figure 2.7). There was
significant variation in average tree carbon content among species in the 12- year-old
plantation (P<0.001).However, there was no significant difference among species in
mean tree carbon content for the 7-year-old plantation (P<0.834). Similarly, carbon
content of Khaya spp grown in wet and moist evergreen forests zones were not
significantly different at age 5.
Apart from Tectona grandis and Cedrella odorata which are exotic species, all others
(i.e. indigenous) tree species were classified into three guilds; pioneers, intermediate and
shade bearers (Hawthorne, 1995). The essence of this classification is to determine if
there is any ecological strategy that might be related to carbon content of tropical tree
species. The result showed no significant difference in carbon content for the guild
classification (Table 2.3).
30
Carbon concentration (%)
50
Bottom
Middle
Top
49
48
47
46
45
44
43
42
Figure 2.1 Carbon concentrations for 12 year-old-trees species from the OCAP plantation
in the wet evergreen forest ecozone of Ghana.
31
Carbon concentration (%)
50
7 years
49
12 years
48
47
46
45
44
Figure 2.2 Comparison of C concentrations for 7 and 12 year-old trees of the same
species from OCAP plantations in the wet evergreen forest ecozone of Ghana.
32
1.6
1.4
Volume (m3)
1.2
1
0.8
0.6
0.4
0.2
0
Figure 2.3 Mean volumes of species in a 12 year-old plantation at OCAP in the wet
evergreen forest ecozone of Ghana.
33
1
a
Wood density (g cm-3)
0.9
0.8
0.7
0.6
0.5
0.4
0.3
bc
b
cde
defg bcd
def bcd
fgh efgh
bc
cdefg gh
efgh defg
Bottom
Middle
Top
defg
defg
h
0.2
0.1
0
Figure 2.4 Wood density estimates of trees species from a 12 year-old plantation at
OCAP in the wet evergreen forest ecozone of Ghana. Bars with different lower case
letters differ significantly (P<0.05).
34
0.6
Wood density (gcm-3)
0.5
0.4
7 years
0.3
12 years
0.2
0.1
0
Heritiera utilis
Khaya ivorensis
Turreanthus africanus
Figure 2.5 Comparison of wood density in trees from 7 and 12 year-old plantations at
OCAP in the wet evergreen forest econzone of Ghana.
35
0.6
Wood density (gcm-3)
0.5
0.4
Moist semi-deciduous ecozone
0.3
Wet evergreen ecozone
0.2
0.1
0
Khaya grandifoliola
Khaya ivorensis
Figure 2.6 Comparison of wood density in 7-year-old Khaya spp from a location at
OCAP in the wet evergreen forest ecozone and Bobiri in the moist semi-deciduous
ecozone of Ghana.
36
200
Carbon content (kgC)
180
160
140
120
100
80
60
40
20
0
Figure 2.7 Mean C content per tree for species growing in a 12 year-old plantation at
OCAP in the wet evergreen forest ecozone of Ghana.
37
Table 2.3
Summary of carbon sequestration for trees species from OCAP plantation in the wet
evergreen and Bobiri in the moist semi-deciduous forests of Ghana.SB (shade earer)
NPLD (non-pioneer light demander), P (pioneers), M (moist semi-deciduous forest),
W (wet evergreen forest).
Age/
location
12 years
12 years
12 years
Guilds Tree species
Tectona grandis
Cedrela odorata
NPLD Heritiera utilis
Entandrophragma
12 years NPLD angolense
12 years NPLD Khaya ivorensis
Pycnanthus
12 years NPLD angolense
12 years NPLD Aningeria robusta
12 years NPLD Antiaris toxicaria
Tieghemelia
12 years NPLD heckelii
12 years P
Ceiba pentandra
Terminalia
12 years P
ivorensis
Terminalia
superba
12 years P
12 years P
Milicia excelsa
12 years P
Lophira elata
Triplochiton
12 years P
scleroxylem
12 years SB
Mammea africana
Guarea
12 years SB
thompsonii
Turreanthus
12 years SB
africanus
7 years
NPLD Khaya ivorensis
7 years
NPLD Heritieria utilis
Turreanthus
7 years
SB
africanus
5years(W) NPLD Khaya spp
5years(M) NPLD Khaya spp
Carbon
Volume Biomass content
(m3)
(kg)
(kgC)
0.304
172
84
0.730
279
134
0.085
39
19
C
(%)
48.9
48.3
48.5
Density
(g cm-3)
0.57
0.38
0.46
46.3
47.2
0.44
0.52
0.032
0.100
14
52
7
25
46.2
48.0
47.4
0.43
0.50
0.36
0.173
0.064
0.216
74
32
77
34
15
36
47.8
46.8
0.58
0.27
0.114
1.404
66
383
32
179
48.0
0.38
0.381
145
70
46.7
46.6
48.4
0.42
0.46
0.76
0.216
0.297
0.040
91
136
31
42
63
15
47.2
0.43
48.6
0.62
0.095
0.231
40
143
19
70
47.3
0.44
0.013
6
3
48.6
46.6
48.7
0.44
0.49
0.44
0.474
0.058
0.060
206
28
27
100
13
13
48.1
46.4
46.0
0.49
0.53
0.41
0.044
0.027
0.020
22
14
8
10
7
4
38
Allometric Relationships
Allometric equations were developed for relationships between tree main stem volume
and diameter at breast height (dbh), and volume and dbh2×height (D2H). The allometric
equations (Table 2.4) explained over 90% of the variation in volume across species and
individual trees and thus could be applied satisfactorily to determine volume of a forest
stand. Log-log relationships between dbh or D2H and volume were also highly
significant (Figures 2.9 and 2.11, P<0.0001). The residual mean square error (RSME),
31% and 18%, were within the range for most allometric equations for the relationships
between volume and dbh and volume and dbh2×height respectively (Navar 2009). The
values for the scaling coefficients, a and b, were highly significant (P>0.0001) for the log
transformed linear regression and the power functions. The scaling coefficient for power
function b, ranged from 2.3 to 2.7 (Table 2.4) as documented for other tropical forests
(Brown 1989; Chave et al. 2005; Navar 2009). These scaling coefficients (a and b) vary
with species, biomass and climate (Baskerville 1965; Zianis and Mencuccini 2004).
The regression analyses further verified the expected strong relationship (r2>0.93)
between tree main stem volume and dbh (Figures 2.8 and 2.9). It is imperative to note
that the relationship between main stem volume and dbh2×height (Figure 2.10 and 2.11)
was even much stronger (r2>0.97) than volume and diameter only (r2>0.93). The
relationship (r2=0.92) between carbon content and dbh (Figure 2.12; r2 >0.92) was
smaller than the other equations (r2>0.93 or 0.97). This is because, several species
consistently had lower carbon content than others for a given size tree. By dividing
species into groups of species with inherently low and high carbon content for a given
diameter, stronger relationships were found (Figures 2.13 to 2.15). This differs from
39
results for volume, where relationships with dbh and D2H held reasonably well across all
species (compare Figures 2.13 to Figures 2.14 and 2.15).
Table 2.4
Regression equations for volume (m ) and biomass (kg) for trees from OCAP area
in the wet evergreen forest zone of Ghana. All equations are in the form M=a(dbh)b
where a and b are scaling coefficients and M is volume or biomass. RMSE is residual
mean square error and r2 is the coefficient of determination.
3
Equations
Ln volume = a+ b ln(dbh)
Ln volume = a+b ln(dbh2×height)
Volume = a(dbh)b
Volume = a(dbh2×height)b
High volume = a(dbh)b
Low volume= a (dbh)b
High biomass = a(dbh)b
Low biomass = a (dbh)b
High carbon content = a(dbh)b
Low carbon content = a (dbh)b
Ln volume = ln a + b ln dbh
Ln volume = ln a + b ln dbh2×height
a
-9.42410
-9.29090
0.00007
0.00008
0.00009
0.00003
0.05100
0.04170
0.02400
0.01990
-9.57700
-9.39910
40
b
2.54
0.91
2.58
0.92
2.53
2.74
2.47
2.38
2.47
2.37
2.58
0.92
r2
0.94
0.98
0.94
0.98
0.94
0.98
0.95
0.97
0.94
0.97
0.94
0.98
RMSE
0.31
0.18
-
2
y = 7E-05x2.5758
R² = 0.9356
1.8
1.6
Volume (m3)
1.4
1.2
1
0.8
0.6
0.4
0.2
0
0
10
20
30
Dbh (cm)
40
50
60
Figure 2.8 Allometric relationship between volume and dbh (power function) for 18 trees
species from 12-year-old plantation at OCAP in the wet evergreen forest ecozone of
Ghana.
41
1
y = 2.5752x - 9.577
R² = 0.9356
Ln Volume (m3)
0
-1
-2
-3
-4
-5
1.8
2.3
2.8
3.3
Ln Dbh (cm)
3.8
4.3
Figure 2.9 Ln- ln relationships between dbh and volume for 18 trees species from a 12year-old plantation at OCAP in the wet evergreen forest ecozone of Ghana.
42
1.8
y = 8E-05x0.9172
R² = 0.9768
1.6
1.4
Volume (m3)
1.2
1
0.8
0.6
0.4
0.2
0
0
10000
20000
D2H
30000
40000
50000
Figure 2.10 Allometric relationship (power function) between volume and dbh2×height
(D2H) for 18 trees species from a 12-year-old plantation at OCAP in the wet evergreen
forest ecozone of Ghana.
43
1
y = 0.9172x - 9.3991
R² = 0.9768
Ln Volume (m3)
0
-1
-2
-3
-4
-5
5
6
7
8
Ln D2H
9
10
11
Figure 2.11 Ln-ln relationship between volume and D2h for18 trees species from a 12
year old plantation at OCAP in the wet evergreen forest ecozone of Ghana.
44
Carbon content (kgC)
250
y = 0.0269x2 + 3.2666x - 33.557
R² = 0.9275
200
150
100
50
0
0
10
20
-50
30
40
50
60
Dbh (cm)
Figure 2.12 Allometric relationship between C content and dbh for 18 trees
species from 12 year-old plantation at OCAP in the wet evergreen forest ecozone
of Ghana.
45
2
1.8
Volume H
1.6
Volume L
Power (Volume H)
Volume (m3)
1.4
Power (Volume L)
y = 9E-05x2.5341
R² = 0.9437
1.2
1
y = 3E-05x2.7434
R² = 0.9828
0.8
0.6
0.4
0.2
0
0
10
20
30
40
50
60
Dbh (cm)
Figure 2.13 Allometric relationships between trees of high volume and dbh, and low
volume and dbh (power function) from OCAP plantation 12 in the wet evergreen forest
ecozone of Ghana.
46
900
Biomass H
800
Biomass L
Biomass (kg)
700
Power (Biomass H)
Power (Biomass L)
600
y = 0.051x2.4683
R² = 0.9459
500
400
300
200
y = 0.0417x2.3752
R² = 0.9721
100
0
0
10
20
30
Dbh (cm)
40
50
60
Figure 2.14 Allometric relationships between trees of high biomass and dbh, and low
biomass and dbh (power function) from OCAP plantation in the wet evergreen forest
ecozone of Ghana.
47
450
400
Stock H
Carbon content (kg)
350
Stock L
Power (Stock H)
300
y = 0.024x2.4744
R² = 0.944
Power (Stock L)
250
200
150
y = 0.0199x2.3724
R² = 0.9712
100
50
0
0
10
20
30
Dbh (cm)
40
50
60
Figure 2.15 Allometric relationships between trees of high C content and dbh, and low C
content and dbh (power function) from OCAP plantation in the wet evergreen forest
ecozone of Ghana.
48
Discussion
Differences in species wood densities, carbon concentrations and carbon contents were
significant for the 12-year-old trees from the wet evergreen forest ecozone. The wood
density and carbon concentration for the bottom, middle and top positions of the main
stem of12-year-old trees were also significant. Wood density of Khaya spp from the wet
evergreen forest ecozone differed significantly from those of the same age in the moist
semi-deciduous forest ecozone. Wood density was negatively correlated with either
height or dbh of the 12-year-old trees. Wood densities of these trees were positively
correlated with carbon concentration.
Regression analysis showed a strong relationship between volume and dbh and volume
and dbh2×height as was expected.
Carbon Concentration
Carbon concentration varied significantly among the eighteen species studied. This
observation was consistent with similar studies by Elias and Potvin (2003) who reported
significant variation in carbon concentration for 32 neo tropical species. The results
showed a range in carbon concentration of 44-49% that was comparable to results found
by other researchers (Kauppi et al. 1995; Elias and Potvin 2003). Among the eighteen
species analyzed, Tectona grandis harbored the highest carbon concentration of 48.9%
which is similar to previous reports (Kraenzel et al. 2003). The variability in carbon
concentration has been attributed to inherent differences in chemical composition, which
depends on type of wood, geographical location, soil and prevailing climatic conditions
49
(Pettersen 1984). For example, softwoods in North America have a higher carbon
concentration than hardwoods because of higher proportion of lignin found in softwood
species (Lamlom and Savidge 2003). Carbon concentration appeared slightly higher for
12 year-old trees than 7 year- old trees, and higher for both middle and top positions of
the main stem. Perhaps age is a factor in carbon concentration of trees as it affects the
amount of juvenile wood and mature wood present in the stem (Lamlom and Savidge
2003). The juvenile woods are characterized by higher concentration of lignin and
extractives than mature woods (Zobel and van Buijtenen 1989; Lamlom and Savidge
2003). This could be responsible for the trend for slightly higher carbon concentration
observed for both the middle and top positions of the tree.
Generally, carbon concentration of trees species has been assumed to be 50% for
conversion of biomass to carbon stock (Matthews 1993). However, this work and others
appears to refute this general notion as the use of this assumed figure could often
overestimate actual wood carbon concentration for many species (Lamlom and Savidge
2003; Wauters et al. 2008). For example, the use of 50% as generic value for carbon
accounting has been noted to potentially cause overestimation of about 2.6% in rubber
trees (Wauters et al. 2008). By contrast, a possible underestimation of 6% was indicated
for carbon content of 50-year-old Pinus pinaster stand using 50% value for carbon
calculation (Bert and Danjon 2006). The observed variation in carbon concentration
suggests that species differences need to be taken into account in order to have accurate
estimates of carbon content of trees (Fukatsu et al. 2008). A strong positive relationship
was observed between carbon concentration and wood density which corroborate
previous studies (Elias and Potvin 2003).
50
Volume and Wood Density
Knowledge of wood density is needed in order to convert the volume of forest trees to
biomass (Brown et al. 1989; Fearnside 1997) which can then be used to estimate carbon
content (Brown and Lugo1982; Brown 1997; Zianis and Mencuccini 2004). Wood
density can vary greatly among species (de Castro et al. 1993; Navar 2009; Henry et al.
2010). Wood density was significantly different for the stem positions used in this study
and generally increased from top to base location of the main stem. This observation has
been noted in similar studies elsewhere (Espinoza 2004; Nogueira 2005). In particular,
mean density of trees in Amazon forests is known to decrease from breast height to the
top of the bole (Nogueira 2005). Wood density also increased from the pith to the bark of
the tree (Wiemann and Williamson 2002). The differences in densities of the main stem
positions observed in this study need to be taken into account in biomass estimation
because these differences could potentially overestimate biomass by about 12% if
densities at only the dbh are used in converting tree volumes into biomass instead of
using the mean of all the three positions. Changes in wood density along the stem may be
due to increasing proportions of juvenile wood from base to the top of the tree (Zobel and
van Buijtenen 1989). These juvenile woods have higher moisture content than matured
wood (Zobel and van Buijtenen 1989). Water content has been found to be negatively
associated with density or specific gravity of tree (Suzuki 1999). This may explain the
lower density at the top of the trees.
It is worth noting that estimates of wood density of species reported in this study were in
agreement with wood density estimates available (Bolza and Keating 1972; Reyes et al.
51
1992). For instance Reyes et al. (1992) estimated density of Ceiba pentandra as 0.26 g
cm-3 and 0.45 g cm-3 for Entandrogma angolense, very close to 0.27 g cm-3 and 0.44 g
cm-3 respectively reported in this study. There was wide variability (0.26 to 0.761g cm-3)
in wood density which has also been confirmed by Suzuki (1999). Chave et al. 2009
reported that, species with thick fibers walls have higher mean wood density than species
with thin fiber walls and this may account for the variation in wood density of the tree
species. Fiber thickness is positively correlated with wood density (McDonald et al.
1995).
Densities of Khaya spp of the same age grown in moist semi-deciduous ecozone were
lower than those grown in the wet evergreen forest ecozone. The wetter location
displayed significantly higher density than the relatively dry location. This finding has
also been observed by other authors who noted that wood density varies from one
location to another (Wiemann and Williamson 2002; Baker et al. 2004). According to
Wiemann and Williamson (2002) density has positive association with precipitation and
this could be responsible for high density for the wetter forest zone. However, Steege and
Hammond (2001) reported that wood density is not correlated with precipitation or soil
fertility.
It is also known that there is often an inverse relationship between wood density and rate
of volume growth (Thomas 1996). This reason may account for both fast and slow
growth rate observed with relatively low and high wood density for Ceiba pentandra and
Guarea Thompsonii respectively. Ceiba pentandra and Guarea Thompsonii may be
adapted to low light regimes with slowing growth and high density. However, the
52
majority of the species were quite unique in their behavior as they exhibited either
relatively high density coupled with fast volume growth or low density and slow volume
growth rate, regardless of whether they were classified as light demanders or shade
bearers.
Carbon Stock and Guilds Classification
Tree carbon content was significantly different among species in the 12 year-old
plantation but did not differ for the guilds classification (pioneer, non-pioneer light
demanders, and shade bearers). Ceiba pentandra had the greatest carbon content,
belonged to the pioneer category, and had the highest growth rate volume and relatively
low density. However, other pioneer species such as Lophira elata and Triplochiton
scleroxylem had very low volume, biomass and carbon content at age 12. These results
are thus only in partial agreement with Redondo-Brenes and Mongnini (2006) who
showed that fast growing trees contain extremely high aboveground biomass, despite
often having lower density. The fast growing species (Ceiba pentandra, Tectona grandis,
Cedrella odorata) seems to accumulate high biomass and carbon due to their life strategy
of rapid rates of photosynthesis and rapid vertical growth (Barnes et al. 1998).
Overall, the results from this study demonstrated that wood density was highly variable
among the tree species (0.26 to 0.76 g cm-3), and this had much greater influence on the
carbon content differences among species than did carbon concentration. Thus, it seems
that species wood density and volume growth rate are the driving factors in the
determination of biomass or carbon content. Plantation development in Ghana has
focused on promoting mixtures of fast growing species primarily to provide raw material
53
for industry. The development of these plantations can play a useful environmental role
in carbon storage (Schroeder 1992; Winjum and Schroeder 1997; Nair et al. 2009).
Allometic Relationships
Site specific equations typically provide the most accurate estimate of forest biomass
(Ketterings et al. 2001; Basuki et al. 2009). For example, aboveground biomass estimated
from data with a local allometric equation from the wet natural forest of Ghana were
significantly different, when similar equations from different regions were applied to the
same data (Henry et al. 2010).
Allometric equations were developed for mixed-plantation forests of the wet evergreen
zone of Ghana. The equation developed looked at the relationship between tree volume
and other easily measured forest inventory data such as height and diameter (Chave et al.
2005). Likewise, relationships between carbon content, biomass, and dbh and height were
established. The strong relationship between either biomass or volume and diameterheight was comparable to similar studies (Brown 1997; Ketterings et al. 2001; Navar
2009; Henry et al. 2010 ). Diameter has been noted as a key parameter used for tree
allometry (Zianis and Mencuccini 2004), but equations based on diameter only are often
site specific. Relationships based on a combination of height and diameters allow
potential application to sites of differing quality (Ketterings et al. 2001). There was
improved accuracy of the various equations when tree height was incorporated which was
consistent with other studies (Henry et al. 2010). Height growth of trees reflects
productivity at a particular site that could be attributed to its specific nutrient and/or
moisture availability. The use of mixed-species equation may be desirable in most cases
54
because it mimics real natural forest situation, where there are several species per unit
area. As many as 250 species per hectare could be found in forests of Ghana (Hall and
Swaine 1981). However, species-specific equations have been reported to provide high
accuracy of biomass estimation (Basuki et al. 2009). The equation developed here could
be applied to plantation grown trees whose diameter range fall between 10cm to 60 cm.
It is important to realize that biomass equations often used for estimating carbon content
across species are not as accurate as volume equations which are largely due to the high
variability among species wood density. The volume equations that combined tree height
and diameter were more accurate than diameter alone. Nevertheless, equations that
consist of diameter only are important because data for forest inventories does not always
include tree height, but virtually always includes diameter. Tree height data are usually
not available because it takes more time to collect tree height data. This could be due in
part to poor light penetration underneath most forest canopy which makes measurement
of top of trees from below the canopy extremely difficult. The volume equations
developed in this study and other similar ones are recommended for use in Ghana and
could be adopted for other tropical countries with similar environments to estimate tree
volume. Carbon contents for the trees can then be obtained by species specific density
data (and percent carbon concentration if available) to the volume estimates.
Conclusion and Recommendations
This research was undertaken to provide information needed for determining the carbon
content of tropical trees grown in forest plantations of Ghana and the surrounding region.
Species differences information necessary for estimating carbon content such as wood
55
density, carbon concentration and volume, were investigated. The results of this study
revealed significant differences among tree species in wood density, carbon concentration
and carbon content at a given age. Wood density varied significantly for three stem
positions (bottom, middle and top) and appeared relatively higher at lower locations on
the stem. It is imperative to reiterate that there was wide variability in wood density
among species which suggests it was a more important factor than carbon concentration
for carbon accounting and that general values should not be applied across all species
when calculated volumes are converted to biomass. Wood densities of Khaya spp grown
in the moist semi-deciduous ecozone were lower than those of the same age grown in wet
forest area. This is likely due to the higher amount of rainfall in the wet forest ecozones.
Wiemann and Williamson (2002) have indicated that density is positively related to
amount of rainfall. However, precipitation has often been found not to have any bearing
with density as discussed earlier (Steege and Hammond 2001). To clarify the cause of the
differences related to ecozone, further research targeted at disentangling the likely
environmental and edaphic factors influencing wood density in Ghana is needed.
Particularly, efforts should be geared towards scrutinizing the effects of soil, rainfall and
elevation on wood density.
Common relationship existed across species between tree dimensions and volume, but
species specific information on wood density was needed for determining carbon content.
To improve accuracy of allometric equations for carbon accounting in Ghana and other
tropical countries, local equations have been suggested as such equations fall within the
range of diameter classes of trees in that particular locality (Brown et al. 1989; Henry et
al. 2010). The various volume equations developed in this research are recommended for
56
use in estimating carbon content of forest plantations in Ghana and if possible can be
adopted for other tropical countries. Development of these volume equation and species
specific wood density and carbon concentration has laid a foundation for improving the
accuracy of carbon accounting in tropical Africa, including Ghana. Still, more research is
needed to investigate other above ground carbon pools, including carbon contained in
leaves and branches, and below ground carbon in fine and coarse roots and soil.
57
Literature Cited
Abebrese MO. 2002. Tropical secondary forest management in Africa: reality and
perspectives. FAO. Available at:
http://www.fao.org/DOCREP/006/J0628E/J0628E53.htm. Accessed August 1,
2011.
Amanor KS. 1997. Collaborative forest management, forest resource tenure and the
domestic economy in Ghana. IRDCurrents. 15:10-16.
Amthor JS. 1995. Terrestrial higher-plant response to increasing atmopheric (CO2) in
relation to the global carbon cycle. Global Change Biology. 1:243-274.
Atjay GL, Ketner P, Duvigneaud P. 1979. Terrestrial primary production and phytomass.
In: Bolin B, Degens ET, Kempe S, editors. The global carbon cycle. New York:
John Wiley and Sons Inc. p. 129-182.
Atuahene SKN. 2001. The forest resource of Ghana and research on Hypsipyla robusta
(Moore) (Lepidoptera: Pyralidae) control in mahogany plantations in Ghana.
Hypsipyla shoot borers in Meliaceae. Proceedings of an International Workshop
held at Kandy, Sri Lanka.
Avery TE, Burkhart HE. 1994. Forest measurements. Forest measurements. 5th edition.
New York: McGraw-Hill, Inc.
Baker TR, Phillips OL, Malhi Y, Almeida S, Arroyo L, Fiore Adi, Erwin T, Killeen TJ,
Laurance SG, Laurance WF, Lewis SL, Lloyd J, Monteagudo A, Neill DA,
Patino S, Pitman NCA, Silva JNM, Vasquez Martinez R, 2004. Variation in
wood density determines spatial patterns in Amazonian forest biomass. Global
Change Biology. 10:545-562.
Barnes BV, Zak DR, Denton SR, Spurr SH. 1998. Forest Ecology. 4th edition. New
York: John Wiley and Sons Inc.
Baskerville GL. 1965. Estimation of dry weight of tree components and total standing
crop in conifer stands. Ecology. 46(6):867-9.
Basuki TM, van Laake PE, Skidmore AK, Hussin YA. 2009. Allometric equations for
estimating the above-ground biomass in tropical lowland Dipterocarp forests.
Forest Ecology and Management. 257(8):1684-1694.
Bert D, Danjon F. 2006. Carbon concentration variations in the roots, stem and crown of
mature Pinus pinaster (Ait.). Forest Ecology and Management. 222(1-3):279-295.
58
Bolza E, Keating WG. 1972. African timbers: the properties, uses and characteristics of
700 species. Melbourne: Division of Building Research,CSIRO.
Brown S, E.Lugo A. 1982. The storage and Production of Organic Matter in Tropical
Forests and Their Role in the Global Carbon Cycle. Biotropica. 14(3):161-187.
Brown S, Gillespie AJR, Lugo AE. 1989. Biomass estimation methods for tropical forests
with applications to forest inventory data. Forest Science. 35(4):881-902.
Brown S. 1997. Estimating biomass and biomass change of tropical forests: a primer.
Rome: Food and Agriculture Organisation.
Castro Fd, Williamson GB, Jesus RMd.1993. Radial variation in the wood specific
gravity of Joannesia princeps: the roles of age and diameter. Biotropica.
25(2):176-182.
Chave J, Andalo C, Brown S, Cairns MA, Chambers JQ, Eamus D, Folster H, Fromard
F, Higuchi N, Kira T, Lescure JP, Nelson BW, Ogawa H, Puig H, Riera B,
Yamakura T. 2005. Tree allometry and improved estimation of carbon stocks and
balance in tropical forests. Oecologia. 145(1):87-99.
Chave J, Coomes D, Jansen S, Lewis SL, Swenson NG, Zanne AE. 2009. Towards a
worldwide wood economics spectrum. Ecology Letters. 12(4):351-366.
Clark DB, Clark DA. 2000. Landscape-scale variation in forest structure and biomass in a
tropical rain forest. Forest Ecology and Management. 137(1):185-198.
Dewar RC, Cannell MGR. 1992. Carbon sequestration in the trees, products and soils of
forest -plantations an analysis using UK examples. Tree Physiology. 11(1):49-71.
Dixon RK, Perry JA, Vanderklein EL, Hiol FH. 1996. Vulnerability of forest resources to
global climate change: Case study of Cameroon and Ghana. Climate Research.
6(2):127-133.
Elias M, Potvin C. 2003. Assessing inter- and intra-specific variation in trunk carbon
concentration for 32 neotropical tree species. Canadian Journal of Forest
Research-Revue Canadienne De Recherche Forestiere. 33(6):1039-1045.
Espinoza JA. 2004. Within-tree density gradients in Gmelina arborea in Venezuela. New
Forest. 28:309-317.
Fearnside PM. 1997. Wood density for estimating forest biomass in Brazilian Amazonia.
Forest Ecology and Management. 90(1):59-87.
FAO. 2005. Global forest resources assessment 2005. Available at:
http://www.fao.org/forestry/fra/fra2005/en/. Accessed August 15, 2011.
59
Fukatsu E, Fukuda Y, Takahashi M, Nakada R. 2008. Clonal variation of carbon content
in wood of Larix kaempferi (Japanese larch). Journal of Wood Science.
54(3):247-251.
Hall JB, Swaine MD. 1976. Classification and ecology of closed-canopy forest in Ghana.
Journal of Ecology. 64(3):913-951.
Hall JB, Swaine MD. 1981. Distribution and ecology of vascular plants in a tropical rain
forest. Forest vegetation in Ghana. Forest vegetation in Ghana. The Hague: Dr W.
Junk Publishers.
Hawthorne WD. 1995. Ecological profiles of Ghanaian forest trees. University of Oxford:
Oxford forestry Institute, department of plant Sciences.
Hawthorne WD, Abu-Juam M. 1995. Forest Protection in Ghana: with particular
reference to vegetation and plant species. Gland Switzerland: International Union
for Conservation of Nature and Natural Resources.
Henry M, Besnard A, Asante WA, Eshun J, Adu-Bredu S, Valentini R, Bernoux M,
Saint-Andre L. 2010. Wood density, phytomass variations within and among
trees, and allometric equations in a tropical rainforest of Africa. Forest Ecology
and Management. 260(8):1375-1388.
Hyvonen R, Persson T, Andersson S, Olsson B, Agren GI, Linder S. 2008. Impact of
long-term nitrogen addition on carbon stocks in trees and soils in northern
Europe. Biogeochemistry. 89(1):121-137.
IPCC. 2007. The physical science basis: contribution of working group 1 to the fouth
assessment report of the IPCC. Available at: http://www.ipcc.ch/. Accessed June
10, 2011.
Jindal R, Swallow B, Kerr J. 2008. Forestry-based carbon sequestration projects in
Africa: potential benefits and challenges. Natural Resource Forum. 32:116-130.
Kauppi PE, Tomppo E, Ferm A. 1995. C and N storage in living trees within Finland
since 1950S. Plant and Soil. 168:633-638.
Ketterings QM, Coe R, Noordwijk Mv, Ambagau Y, Palm CA. 2001. Reducing
uncertainty in the use of allometric biomass equations for predicting aboveground tree biomass in mixed secondary forests. Forest Ecology and
Management. 146(1):199-209.
Kraenzel M, Castillo A, Moore T, Potvin C. 2003. Carbon storage of harvest-age teak
(Tectona grandis) plantations, Panama. Forest Ecology and Management. 173(13):213-225.
60
Lamlom SH, Savidge RA. 2003. A reassessment of carbon content in wood: variation
within and between 41 North American species. Biomass and Bioenergy.
25(4):381-388.
Lamlom SH, Savidge RA. 2006. Carbon content variation in boles of mature sugar maple
and giant sequoia. Tree Physiology. 26:459-468.
Litton CM, Ryan MG, Knight DH. 2004. Effects of tree density and stand age on carbon
allocation patterns in postfire lodgepole pine. Ecological Applications. 14(2):460475.
Luizao RCC, Luizao FJ, Paiva RQ, Monteiro TF, Sousa LS, Kruijt B. 2004. Variation of
carbon and nitrogen cycling processes along a topographic gradient in a central
Amazonian forest. (The Large Scale Biosphere-Atmosphere Experiment in
Amazonia - LBA). Global Change Biology. 10(5):592-600.
Mahlman JD. 1997. Uncertainties in the projections of human-caused climate warming.
Science. 278:1416-1417.
Malhi Y, Meir P, Brown S. 2002. Forests, carbon and global climate. Philosophical
Transactions of the Royal Society of London Series a-Mathematical Physical and
Engineering Sciences. 360(1797):1567-1591.
Matthews G. 1993. The carbon content of trees. Edinburgh: Technical Paper-UK Forestry
Commission.
McDonald SS, Williamson GB, Wiemann MC. 1995. Wood specific gravity and anatomy
in Heliocarpus appendiculatus (Tiliaceae). American Journal of Botany.
82(7):855-861.
Nair PKR, Kumar BM, Nair VD. 2009. Agroforestry as a strategy for carbon
sequestration. Journal of Plant Nutrition and Soil Science-Zeitschrift Fur
Pflanzenernahrung Und Bodenkunde. 172(1):10-23.
Nanang DM. 2010. Analysis of export demand for Ghana's timber products: a
multivariate co-integration approach. Journal of Forest Economics. 16(1):47-61.
Navar J. 2009. Allometric equations for tree species and carbon stocks for forests of
northwestern Mexico. Forest Ecology and Management. 257(2):427-434.
Nogueira EM, Nelson BW, Fearnside PM. 2005. Wood density in dense forest in central
Amazonia, Brazil. Forest Ecology and Management. 208:261-286.
Parresol BR. 1999. Assessing tree and stand biomass: a review with examples and critical
comparisons. Forest Science. 45(4):573-593.
61
Pettersen RC. 1984. The chemical-composition of wood. Advances in Chemistry Series.
(207):57-126.
Redondo-Brenes A, Montagnini F. 2006. Growth, productivity, aboveground biomass,
and carbon sequestration of pure and mixed native tree plantations in the
Caribbean lowlands of Costa Rica. Forest Ecology and Management. 232:168178.
Reyes G, Brown S, Chapman J, Lugo AE. 1992. Wood densities of tropical tree species.
General Technical Report SO-88 - Southern Forest Experiment Station, USDA
Forest Service.
Sandker M, Nyame SK, Forster J, Collier N, Shepherd G, Yeboah D, Ezzine-de Blas D,
Machwitz M, Vaatainen S, Garedew E,Etoga G, Ehringhaus C, Anati J, Quarm
ODK, Campbell BM. 2010. REDD payments as incentive for reducing forest loss.
Conservation Letters. 3(2):114-121.
SAS Institute. 1997. SAS/STAT user’s guide. 4th edition. SAS institute: Cary, NC.
Schneider SH.1990. The global warming debate heatss up: an analysis and perspective.
Bullectin of the American Meterological Society. 71:1292-1304.
Schroeder P. 1992. Carbon storage potential of short rotation tropical tree plantations.
Forest Ecology and Management. 50(1-2):31-41.
Sicard C, Saint-Andre L, Gelhaye D, Ranger J. 2006. Effect of initial fertilisation on
biomass and nutrient content of Norway spruce and Douglas-fir plantations at the
same site. Trees: Structure and Function. 20(2):229-246.
Slik JWF, Bernard CS, Breman FC, Beek Mv, Salim A, Sheil D. 2008. Wood density as a
conservation tool: quantification of disturbance and identification of conservationpriority areas in tropical forests. Conservation Biology. 22(5):1299-1308.
Steege Ht, Hammond DS. 2001. Character convergence, diversity, and disturbance in
tropical rain forest in Guyana. Ecology. 82(11):3197-3212.
Suzuki E. 1999. Diversity in specific gravity and water content of wood among Bornean
tropical rainforest trees. Ecological Research. 14(3):211-224.
Thomas SC. 1996. Asymptotic height as a predictor of growth and allometric
characteristics in Malaysian rain forest trees. American Journal of Botany.
83(5):556-566.
62
UNFCCC. 1997. Clean development mechanism. Available at:
http://unfccc.int/kyoto_protocol/mechanism/clean_development_mechanism/item
s/2718.php. Accessed August 12, 2011.
Wauters JB, Coudert S, Grallien E, Jonard A, Ponette Q. 2008. Carbon stock in rubber
tree plantations in Western Ghana and Mato Grosso (Brazil). Forest Ecology and
Management. 255(7):2347-2361
.
Winjum JK, Schroeder PE. 1997. Forest plantations of the world: Their extent, ecological
attributes, and carbon storage. Agricultural and Forest Meteorology. 84(1-2):153167.
Wirth C, Schumacher J, Schulze ED. 2004. Generic biomass functions for Norway spruce
in Central Europe - a meta-analysis approach toward prediction and uncertainty
estimation. Tree Physiology. 24(2):121-139.
Zianis D, Mencuccini M. 2004. On simplifying allometric analyses of forest biomass.
Forest Ecology and Management. 187(2):311-332.
Zobel BJ, Buijtenen JPv. 1989. Wood variation: its causes and control. Berlin: SpringerVerlag.
63