Richness of Forest Stands and Atmospheric Carbon Dioxide Storage in Urban Institutional Lands of Bukavu, D.R. Congo

Abstract

Improving the urban environmental quality relies mainly on the increasing of urban forests capacity to store carbon dioxide. This study assesses the floristic diversity of urban institutional lands in Bukavu and their potential to reduce atmospheric CO2. An exhaustive inventory over three sites (Collège Alfajiri, Cathédrale Notre-Dame de la Paix and Institut Supérieur Pédagogique) of Bukavu led to the identification of 1 113 trees of which the diameter at breast height (1.30 m) ranged from 4.9 to 161 cm. Results reveal a floristic diversity made up of 4 families of conifers with 4 species and 14 of broadleaves with 21 species. Average densities were of 54 trees ha-1 and 5.21 m2 ha-1 of basal area. Urban-based allometric equations used yielded up to 312.8 tons of carbon stored in trees aboveground biomass equivalent to 1 147.9 tons of CO2 reduced from the atmosphere over the three sites. The rate of carbon storage reaches 15.1 tons ha-1. Thus, trees of the three institutional sites in Bukavu play an important role in reducing atmospheric CO2 and contribute, thereby, to mitigate global climate change effects. Given the current environmental challenge associated with high population growth rate in cities, the urban forest ecosystem in DRC requires to be extended and further
investigation.

Full text

Regular Article
J
F
E
S
Journal of Forest and
Environmental Science
pISSN: 2288-9744, eISSN: 2288-9752
J
ournal of Forest and Environmental Science
V
ol. 33, No. 2, pp. 79-90, May, 2017
https://doi.org/10.7747/JFES.2017.33.2.79
J For Environ Sci 33(2), 79-90 79
Richness of Forest Stands and Atmospheric
Carbon Dioxide Storage in Urban Institutional
Lands of Bukavu, D.R. Congo
Bakach D. KADIATA1,* and J. B. Ncutirakiza NDAMIYEHE
2
1Faculté des Sciences Agronomiques, Université de Kinshasa, B.P. 117, Kinshasa XI, Democratic Republic of Congo
2
Faculté des Sciences, Université de Kisangani, B.P. 2012 Kisangani, Democratic Republic of Congo
Abstract
Improving the urban environmental quality relies mainly on the increasing of urban forests capacity to store carbon
dioxide. This study assesses the floristic diversity of urban institutional lands in Bukavu and their potential to reduce
atmospheric CO2. An exhaustive inventory over three sites (Collège Alfajiri, Cathédrale Notre-Dame de la Paix and
Institut Supérieur Pédagogique) of Bukavu led to the identification of 1,113 trees of which the diameter at breast height
(1.30 m) ranged from 4.9 to 161 cm. Results reveal a floristic diversity made up of 4 families of conifers with 4 species
and 14 of broadleaves with 21 species. Average densities were of 54 trees ha-1 and 5.21 m2 ha-1 of basal area. Urban-based
allometric equations used yielded up to 312.8 tons of carbon stored in trees aboveground biomass equivalent to 1,147.9
tons of CO2 reduced from the atmosphere over the three sites. The rate of carbon storage reaches 15.1 tons ha-1. Thus,
trees of the three institutional sites in Bukavu play an important role in reducing atmospheric CO2 and contribute,
thereby, to mitigate global climate change effects. Given the current environmental challenge associated with high
population growth rate in cities, the urban forest ecosystem in DRC requires to be extended and further investigation.
Key Words: aboveground biomass, allometric equations, broadleaves/conifers, carbon dioxide, urban forest
Received: May 12, 2016. Revised: November 18, 2016. Accepted: November 20, 2016.
Corresponding author: Bakach D. KADIATA
Faculté des Sciences Agronomiques, Université de Kinshasa, B.P. 117, Kinshasa XI, Democratic Republic of Congo
Tel: 243 998 92 32 90, Fax: 243 815 560 294, E-mail : dikand56@gmail.com; bakach_kadiata@yahoo.c om
Introduction
Cities represent an environmental challenge due to the
high levels of air pollutants but also to the overwhelming
temperatures (Vergriette and Labrecque 2007), and multi-
ple artificial surfaces which expose them exceedingly to the
consequences of global climate change (Nowak and Crane
2002). Fortunately the forest vegetation particularly plays a
key role in improving the quality of the urban environment
(Yang et al. 2005; Nowak et al. 2008). In fact, trees shade
surfaces and reflect sunlight to reduce temperatures; cool
local air through evapotranspiration; supply citizens with
oxygen, reduce air pollutants by capturing particulate mat-
ters, absorbing gaseous pollutants like sulfur dioxide (SO2),
nitrogen dioxide (NO2) and ozone (O3) (Peck et al. 1999;
Yang et al. 2005; Bell and Wheeler 2006; Lessard and
Boulfroy 2007).
In addition, urban trees offer a double benefit in re-
ducing atmospheric CO2. First, they directly store and se-
quester atmospheric carbon while they grow. Second, prop-
erly located around buildings, trees in urban areas also con-
serve energy by reducing the demand for heating and air
conditioning, thereby reducing emissions associated with
electricity power production (Nowak 1993; Nowak 1994a;

Urban Forest Cover and Carbon Dioxide Storage
80 Journal of Forest and Environmental Science http://jofs.or.kr
McPherson 1998; Brack 2002). Thus, urban forest man-
agement programs for improving the quality of the urban
environment are, in many ways, necessary.
However, in the Democratic Republic of Congo (DRC),
data on urban trees or forest ecosystems remain rather scan-
ty, yet they represent useful information to undertake man-
agement of the urban forest cover and could serve as a tool
basis with which to plan actions for enhancing and main-
taining the ligneous potential in overpopulated cities.
Therefore, there is a dearth of documented information
on numerous benefits provided by the vegetation, especially
the amount of carbon sequestered by trees in urban areas of
DRC. Yet, it would help to assess the contribution of cities
in mitigating climate change effects and, hence, serve to
claim the offset in the international system of carbon
credits. The improvement of carbon sinks as of many other
tree functions faces, in cities of DRC, the constraint due to
lack of baseline research in urban forestry field. Then, the
general lack of data on trees in cities arouses the interest to
undertake, nowadays, studies on the urban forest ecosystem
of DRC including the appraisal of its ability to store atmos-
pheric carbon dioxide.
Thus, this study aims to (1) assess the structure of trees
including species composition identification, tree size and
density in three of institutional sites in Bukavu and (2) to
determine how the amount of carbon stored from the at-
mosphere does vary within these sites.
Materials and Methods
Study area
The study was conducted in Bukavu, an eastern city of
the Democratic Republic of Congo being the capital of the
South-Kivu province. Administratively, it covers 60 km2
and is composed of four municipalities, which are Ibanda,
Kadutu, Bagira and Kasha making up to 1 million in-
habitants in 2014.
Located at about 1,600 m on the southwestern shore of
the Lake Kivu, Bukavu lies between 2o 26' and 2o 33' south
and 28o 49' and 28o 53' East (Anonymous 2012). The land
is very uneven and increases gradually as one moves away
from the Lake. The climate is humid temperate with the
altitude. The rainfall ranges from 1,500 to 2,000 mm per
year while the annual average temperature is around 19°C.
The two seasons in Bukavu are unequally distributed over
the year; the dry one is spread over 3 months from June to
September, while the rainy season covers the remaining 9
months. The soil of Bukavu, of volcanic type, relatively fer-
tile, clayey and permeable, belongs entirely to the group of
reddish clay soils (Anonymous 2005; Anonymous 2012;
Polepole 2007).
Study sites
For selecting institutional sites in which this study was
carried out, the physical characteristics of the sites, their lo-
cation and the accessibility for data collection were the most
important criteria. Thus, three public institutional sites
were selected in the city of Bukavu, and identified as S-ISP
for Institut Supérieur Pédagogique de Bukavu, S-CA for
Collège Alfajiri and S-ND for Cathédrale Notre-Dame de
la Paix. Diverse in their location in the town landscape, in
land use characteristics as well as in their property status as
being a church, a secondary and a higher educational lands,
these sites constitute a sizeable and reliable sample for such
investigation.
Materials
Three types of materials used comprised of biological
ones consisting of trees on which measurements were taken,
and the various key tools used for data collection on the
ground among which a Global Positioning System (Gar-
minTM GPS) and a diameter tape of 6.5 m. The third cat-
egory consisted of tools for data processing notably, Micro-
soft Excel software and the allometric equations as well.
Tree inventory
All trees in the three institutional sites selected in Bukavu
were inventoried and thereafter subjected to species identi-
fication, diameter measurements, site area calculation and
tree coordinates determination. The area determination of
each of the three study sites was conducted in activating the
GPS option related to ‘area calculation’ and turning around
the considered site. Then, the GPS served for locating each
of the trees, using three variables which are latitude, longi-
tude and elevation above sea level. The identification of
some species was facilitated by thereon pasted labels with
scientific names and botanical families of trees existing
within the S-ISP. For all species, the expertise of botanists

Bakach D. KADIATA and
J
. B. Ncutirakiza NDAMIYEH
E
J For Environ Sci 33(2), 79-90 81
Ta b le 1. Individual and generic allometric equations used*
DBH only DW Density for
Vo l . t o DW c on ve r si on
Species DBH Range (cm) Volume (m3)(kg/m
3)
Acacia longifolia 15.0-57.2 0.0283168466(0.048490*(dbh/2.54)2.347250)630
Cupressus macrocarpa 15.7-146.6 0.0283168466(0.035598*(dbh/2.54)2.495263)460
Eucalyptus globulus 15.5-130 0.0283168466(0.055113*(dbh/2.54)2.436970)620
Jacaranda mimosifolia 17.3-59.7 0.0283168466(0.036147*(dbh/2.54)2.486248)380
Pinus radiata 16.8-105.4 0.0283168466(0.019874*(dbh/2.54)2.66079)440
General Broadleaf 6.4-136.7 0.280285*(dbh cm)⌃2.310647 Multiply FW by 0.56
General Conifer 6.4-136.7 0.05654*(dbh cm)⌃2.580671 Multiply FW by 0.48
*Source: Center for Urban Forest Research (2008); Bruyat (2011).
was secured for identification.
The diameter at breast height (DBH) at 1.30 m was
measured on all trees inventoried. In addition, information
on the morphological structure of each tree was noted.
Data analysis
Results from angiosperms and gymnosperms were sepa-
rated during data processing. However, gymnosperms are
often designed as conifers because all gymnosperms trees
inventoried in Bukavu belong to the coniferous group. All
the same, angiosperms are designed as broadleaves or
hardwoods.
Tree size distribution
The structure of trees diameter is determined to charac-
terize the size distribution of trees in order to assess the in-
tensity of the planting activity in recent years and their po-
tential for CO2 sequestration. Thus, nine diameter classes
of 10-cm interval were defined using Sturge’s formula :
Number of classes=1＋3.3 log n; with n being sample size.
Tree density calculation
To express the tree cover within chosen institutional sites,
the density was calculated in terms of (a) number of trees
per site area as well as of (b) basal stem area expressing the
percentage covered by tree canopies over the ground area
they occupy, as follows:
(a) Density (N/ha)=Number of trees of all species/Site
area (ha)
(b) Basal area (m2/ha)=
∑












   
where i is an integer varying from 1 to n stems
Di diameter at breast height (1.30 m) for i stem
Carbon stock estimates
Current allometric equations were used to calculate trees
carbon storage. They are from 26 open-grown trees equa-
tions used in the Tree Carbon Calculator (CTCC) and de-
scribed as typical of urban areas (CFUR 2008; Bruyat
2011; Aguaron and McPherson 2011). Proposed by the
Center for Urban Forest Research (CUFR 2008), a
United States research laboratory specialized in urban for-
estry, these equations compute aboveground volume of
trees based on their DBH. However, specific equations for
some of surveyed species in the landscape of Bukavu were
not found. In such cases, allometric equations of trees be-
longing to the same genus were used; otherwise generic
equations of conifers and broadleaves were applied. Table 1
lists different specific and generic equations used in this
study. Thus, to determine the total carbon stored, the dry
weight biomass was first calculated.
Dry weight biomass calculation
Generic allometric formulas for broadleaves and conifers
(Table 1) served in calculating directly the fresh weight bio-
mass of trees. However, individual equations of Table 1
computed the green volume of trees. In both cases, con-
version factors were used to derive the dry weight biomass.
Thus, for specific equations, conversion factors in the

Urban Forest Cover and Carbon Dioxide Storage
82 Journal of Forest and Environmental Science http://jofs.or.kr
Ta b le 2. Distribution of trees in institutional sites by species and family
Family Species
Number of trees per institutional site
To t a l
S-ND S-ISP S-CA
Conifers Araucariaceae Araucaria araucania 0101
Cupressaceae Cupressus lusitanica 74 5 41 120
Pinaceae Pinus patula 310031
Podocarpaceae Podocarpus usambarensis 162523
Total conifers 121 8 46 175
Broadleaves Anacardiaceae Mangifera indica 0202
Agavaceae Dracaena steudneri 0011
Annonaceae Annona reticulata 0112
Apocynaceae Plumeria alba 0202
Asteraceae Vernonia amygdalina 0033
Bignoniaceae Jacaranda mimosifolia 0 19 111 130
Markhamia lutea 25 3 52 80
Spathodea campanulata 531725
Combretaceae Terminalia chebula 37 14 0 51
Euphorbiaceae Euphorbia candelabrum 0303
Euphorbia tirucalli 0101
Fab a ce a e A c ac i a s pp 0453580
Albizia lebbeck 0606
Albizia gummifera 005555
Leucaena leucocephala 02911
Lauraceae Persea americana 2428
Meliaceae Cedrela serrata 28 43 13 84
Melia azedarach 23 2 14 39
Moraceae Ficus exasperata 001111
Myrtaceae Eucalyptus globulus 0 232 11 243
Proteacea Grevillea robusta 82667101
Total broadleaves 128 408 402 938
Total trees 249 416 448 1,113
column ‘DW density for volume to DW conversion’ of
Table 1 allow to move from volume to dry weight biomass
as follows:
Dry weight biomass (kg)=Vo l um e (m3)*DW density
DW density is dry weight density expressed in kg/m3
However, generic equations computing the fresh weight
(FW) biomass were multiplied by the conversion factors of
0.48 and 0.56 to obtain dry weight biomass respectively for
conifers and broadleaves.
Carbon stock and CO2 calculation
Dry weight biomass of trees was converted into total car-
bon stock by multiplying its value by 0.5 according to
Nowak et al. (2008) and Nowak and Crane (2002). To ob-
tain the amount of atmospheric CO2, values of carbon stock
were multiplied by 3.67, the molecular weight of carbon di-
oxide, according to McPherson (1998). Then data of car-
bon dioxide stock were reported in metric tons by multi-
plying their values expressed in kilograms by 0.001.
Error estimate
Equations of Table 1 were designed for being used with
a DBH at 4.5 ft (1.37 m). But, as diameters used in this
study were measured at 1.30 m, volumes and biomass cal-
culated tend to be overestimated. However, given that such
overestimation accounted for less than 1% (Bruyat 2011),

Bakach D. KADIATA and
J
. B. Ncutirakiza NDAMIYEH
E
J For Environ Sci 33(2), 79-90 83
Fig. 1 . Distribution of trees into
diameter classes, Bukavu (October
2013).
the gap was found negligible.
Results
Species composition
Trees surveyed in institutional urban sites of Bukavu are
of diverse nature and of species belonging to numerous dif-
ferent botanical families within the two classes, angiosperms
and gymnosperms. A detailed breakdown of inventoried
trees within the three study sites (Table 2) reveals a wide
tree diversity characterized by 4 species representing 4 fam-
ilies of conifers and 21 species of 14 families of broadleaves
(Table 2). In contrast to scarcity of Araucaria araucania
which was only found in the ISP site and Pinus patula pres-
ent only in S-ND, other species of gymnosperms (Cupressus
lusitanica and Podocarpus usambarensis) are rather common
to all the three sites.
Among identified angiosperms only five species, namely
Markhamia lutea, Spathodea campanulata, Persea ameri-
cana, Melia azedarach and Grevillea robusta are widespread
in the three sites of Bukavu. Nevertheless, only four fami-
lies i.e. Bignoniaceae, Euphorbiaceae, Fabaceae and Melia-
ceae were represented by more than one species.
Among sites, numbers of broadleaves and conifers trees
varied greatly (Table 2). On the overall, stems of broad-
leaves are by far more numerous than those of conifers.
Thus, the ratio broadleaves/conifers are 1.05; 8.74 and 51
respectively in S-ND, S-CA and S-ISP.
Tree distribution into diameter classes
Fig. 1 displays the number of trees per diameter class for
both conifers and broadleaves. For conifers, diameter
classes of 20-30 cm and 30-40 cm show the highest num-
bers and are followed by 10-20 cm and 40-50 cm classes.
These four diameter classes comprise 85% of all trees
surveyed. Specifically, more than 80% of conifers have a
DBH between 10 and 50 cm (Fig. 1). Meanwhile, the
10-20 cm up to 40-50 cm diameter classes are the most
dominant for broadleaves, making up nearly 84% of total
inventoried trees. Thus, it is particularly observed that for
each class size, broadleaves remain by far more prevalent
than conifers. Irrespective of study sites, the number of
trees shows a decreasing pace with increasing diameter
girth as from the 10-20 cm class (Fig. 1).
Tree density in institutional lands
Total area calculated of all study sites reaches 20.7 ha of
which 11.4 ha belong to Collège Alfajiri (S-CA), 2 ha to
Cathédrale Notre-Dame de la Paix (S-ND) and 7.3 ha to
Institut Supérieur Pédagogique de Bukavu (S-ISP). These
values were useful for estimating tree density in each site as
shown in Table 3.

Urban Forest Cover and Carbon Dioxide Storage
84 Journal of Forest and Environmental Science http://jofs.or.kr
Ta b le 3. Tree density and basal area by species within institutional lands
Density (trees/ha) Basal area (m2/ha)
Conifers Araucaria araucania 0.0 0.14 0.0 0.05 0.00 0.02 0.00 0.00
Cupressus lusitanica 37 0.68 3.6 5.80 5.06 0.08 0.35 0.71
Pinus patula 15.5 00 00 1.50 1.30 0.0 0.0 0.13
Podocarpus usambarensis 8.0 0.27 0.44 1.11 1.14 0.02 0.24 0.25
Broadleaves Albizia gummifera 0.0 0.0 4.82 2.66 0.00 0.0 0.54 0.30
Albizia lebbeck 0.0 0.82 0.0 0.29 0.00 0.01 0.0 0.00
Acacia spp 0.06.163.073.860.000.1 0.240.17
A. reticulata 0.00.140.090.100.000.0 0.0 0.00
Cedrela serrata 14.0 5.89 1.14 4.06 1.17 0.08 0.09 0.19
Dracaena steudneri 0.0 00 0.09 0.05 0.00 0.0 0.0 0.00
Eucalyptus globulus 0.0 31.78 0.96 11.74 0.00 4.96 0.13 1.82
Euphorbia candelabrum 0.0 0.41 0.0 0.14 0.00 0.03 0.0 0.01
Euphorbia tirucali 0.0 0.14 0.0 0.05 0.00 0.01 0.0 0.00
Ficus exasperata 0.0 00 0.96 0.53 0.00 0.00 0.25 0.14
Grevillea robusta 4.0 3.56 5.88 4.88 0.28 0.37 0.4 0.38
Jacaranda mimosifolia 0.0 2.6 9.74 6.28 0.00 0.08 0.59 0.35
Leucaena leucocephala 0.0 0.27 0.79 0.53 0.00 0.02 0.19 0.11
Mangifera indica 0.0 0.27 0.0 0.10 0.00 0.01 0.0 0.00
Markhamia lutea 12.5 0.41 4.56 3.86 0.38 0.00 0.18 0.13
Melia azedarach 11.5 0.27 1.23 1.88 2.07 0.00 0.20 0.31
Persea americana 1.0 0.55 0.18 0.39 0.06 0.10 0.01 0.05
Plumeria alba 0.0 0.27 0.0 0.10 0.00 0.00 0.00 0.00
Spathodea campanulata 2.5 0.41 1.49 1.21 0.20 0.02 0.07 0.07
Terminalia chebula 18.5 1.92 0.0 2.46 0.72 0.02 0.00 0.08
Vernonia amygdalina 0.0 0.0 0.26 0.14 0.00 0.00 0.01 0.01
Total 124.5 56.96 39.30 53.77 12.38 5.93 3.49 5.21
Thus, the tree density both in terms of number per ha as
well as of basal area varies widely among species and from
one site to another in the city of Bukavu (Table 3).
In terms of number of trees per area, a wide variability
prevailed among species to such an extent that it ranged
from 0.0 to 31.8 trees ha-1. In the S-ND, with a mean of
125 trees ha-1, the most dominant species are Cupressus lu-
sitanica and Te r m i n a l i a c h e b u l a . Eucalyptus globulus is the
species most represented in S-ISP with about 32 trees ha-1,
while Jacaranda mimosifolia and Grevillea robusta are den-
sest in S-CA where the average is of 39 trees ha-1.
The largest basal area up to 12.38 m2 ha-1 is found in the
S-ND against half of it in the S-ISP and about 3.6 times
lesser in the S-CA. Furthermore, the most dominant spe-
cies in stems per ha were not necessarily covering the great-
est ground area. This situation is shown in Table 4 catego-
rizing species according to the descending order of their
relative density. The most abundant species is Eucalyptus
globulus making up to 21.8% of total trees surveyed on
each hectare of institutional site, then follows Jacaranda
mimosifolia with 11.7% on average over all sites (Table 4).
Moreover, beyond the average of 5.21 m2 ha-1 of basal area
occupied by trees in institutional lands, 34.93% of this
ground area is covered by Eucalyptus globulus followed by
Cupressus lusitanica with 13.6% (Table 4). Besides, species
such as Albizia gummifera, Melia azedarach and Podocar -
pus usambarensis outpace by their percent of basal area oth-
ers such as Acacia spp, Cedrela serrata and Markhamia lu-
tea that ranked among the seven most abundant species per
hectare.

Bakach D. KADIATA and
J
. B. Ncutirakiza NDAMIYEH
E
J For Environ Sci 33(2), 79-90 85
Ta b le 4. The most dominant species by percent of density as trees/ha and as basal area
N0Tree density Basal area
Species % Species %
1Eucalyptus globulus 21.83 Eucalyptus globulus 34.93
2Jacaranda mimosifolia 11.68 Cupressus lusitanica 13.63
3Cupressus lusitanica 10.78 Grevillea robusta 7.29
4Grevillea robusta 9.07 Jacaranda mimosifolia 6.72
5Cedrela serrata 7.55 Melia azedarach 5.95
6Acacia spp 7.19 Albizia gummifera 5.76
7Markhamia lutea 7.19 Podocarpus usambarensis 4.80
8 Other 24.71 Other 20.92
Total 100.00 Total 100.00
Ta b le 5. Total carbon stored and CO2 amount by species within urban sites of Bukavu
Espèces
S-ISP S-ND S-CA Total
Carbon (t) Carbon (t) Carbon (t) Carbon (t) CO2 (t)
Conifers Araucaria araucania 0.16 0.00 0.00 0.16 0.57
Cupressus lusitanica 1.20 20.45 7.76 29.40 107.92
Pinus patula 0.00 3.60 0.00 3.60 13.22
Podocarpus usambarensis 0.19 4.45 6.50 11.14 40.89
Total conifers 1.55 28.50 14.26 44.31 162.60
Broadleaves Acacia spp 0.77 0.00 5.83 6.60 24.24
Albizia gummifera 0.00 0.00 19.61 19.61 71.98
Albizia lebbeck 0.08 0.00 0.00 0.08 0.31
A. reticulata 0.05 0.00 0.04 0.09 0.34
Cedrela serrata 1.38 7.12 2.96 11.45 42.04
Draceana steudneri 0.00 0.00 0.11 0.11 0.40
Euphorbia candelabrum 0.56 0.00 0.00 0.56 2.04
Eucalyptus globulus 128.70 0.00 5.22 133.92 491.50
Euphorbia tirucali 0.28 0.00 0.00 0.28 1.03
Ficus exa spera ta 0.00 0.00 12.23 12.23 44.88
Grevillea robusta 8.68 1.65 13.41 23.73 87.11
Jacaranda mimosifolia 0.57 0.00 9.18 9.75 35.78
Leucaena leucocephala 0.32 0.00 8.29 8.62 31.62
Mangifera indica 0.21 0.00 0.00 0.21 0.76
Markhamia lutea 0.01 1.96 5.49 7.47 27.40
Melia azedarach 0.00 14.01 7.64 21.66 79.48
Persea americana 2.74 0.34 0.00 3.09 11.33
Plumeria alba 0.07 0.00 0.22 0.29 1.06
Spathodea campanulata 0.40 1.21 2.57 4.18 15.33
Terminalia chebula 0.33 3.91 0.00 4.24 15.57
Vernonia amygdalina 0.00 0.00 0.32 0.32 1.16
Total broadleaves 145.16 30.20 93.13 268.49 985.35
Total 146.71 58.70 107.39 312.79 1,147.95

Urban Forest Cover and Carbon Dioxide Storage
86 Journal of Forest and Environmental Science http://jofs.or.kr
Fig. 2 . Tons of carbon stored per
hectare in institutional urban sites
of Bukavu (October 2013).
Species distribution in relation with allometric equa-
tions used
Total carbon was computed for 1,027 trees amounting to
92% of the whole sample. Thus, specific allometric equa-
tions of Eucalyptus globulus and Jacaranda mimosifolia
were used to calculate carbon content in 320 trees, corre-
sponding to 31% of all inventoried trees. Besides, allo-
metric equations of Acacia longifolia, Pinus radiata and
Cupressus macrocarpa allowed estimates of carbon fixed in
198 trees of their congeners, whose proportions are re-
spectively 49 stems of Acacia spp, 119 of Cupressus lusitan-
ica and 30 of Pinus patula. At last, trees for which generic
equations of conifers and broadleaves were applied repre-
sent 49.5% of all trees surveyed, respectively 24 stems of 2
conifers species and 485 stems of 18 broadleaves.
Carbon stocks distribution per institutional sites and
species
Carbon sequestered in the aboveground biomass of coni-
fers and broadleaves and the amount of CO2 fixed from the
atmosphere in each institutional site are detailed in Table 5.
Data reveal a wide variability of both parameters values
within and among institutional sites. For the three study
sites (S-ISP, S-ND and S-CA), the total carbon stored
amounted to 312.8 tons, being equivalent to 1,147.9 tons of
CO2 fixed from the atmosphere (Table 5).
Likewise, the amount of carbon stored differs from one
species to another due to their own characteristics, among
which the DBH remains important (Table 5). Thus, the
amount of carbon stored at the ISP site is the highest of all
study sites and is more than twice that in S-ND. Species
holding more carbon at S-ISP are Cupressus lusitanica
with 77.5% of total stock of conifers and Eucalyptus glob-
ulus with 88.6% of total carbon fixed by broadleaves.
Furthermore, in the S-ND, 70.8% of total carbon is stored
by three species, namely Cupressus lusitanica, Cedrela ser-
rata and Melia azedarach.
Fo r Collège Alfajiri site, the equivalent of total CO2 fixed
from the atmosphere by conifers amounted to 52.3 tons
mainly from two species, Cupressus lusitanica and Podoc ar pus
usambarensis. However, broadleaves reveal a potential of
carbon of nearly 93.1 tons equivalent to 341.8 tons of CO2
(Table 5). Species with the greatest stocks in S-CA are
Albizia gummifera, Ficus exasperata and Grevillea robusta
holding 42.1% of total carbon storage within.
Overall, compared on the basis of carbon storage, coni-
fers and broadleaves were similar in the S-ND. Elsewhere,
broadleaves were by far superior. Thus, carbon storage ra-
tios of broadleaves/conifers are 1.06, 6.53 and 93.6 re-

Bakach D. KADIATA and
J
. B. Ncutirakiza NDAMIYEH
E
J For Environ Sci 33(2), 79-90 87
Fig. 3 . Partitioning of carbon fixed
by the most efficient species in ur-
ban area of Bukavu (October 2013).
spectively in the S-ND, S-CA and S-ISP.
Carbon density variation among sites
Total carbon stored per hectare widely varies from one
institutional site to another in the city of Bukavu (Fig. 2).
The greatest amount of carbon stored per ha occurs in the
S-ND (29.3 tons C ha-1) and is three-fold higher than that
of the Collège Alfajiri site. For the three sites, the average is
almost 15.1 tons of carbon accumulated per ha of institu-
tional area.
Highest carbon sink species
The percent of carbon stored by species classified the
most dominant is shown in Fig. 3. It is observed that the
seven most successful species on all sites belong to both
conifers and broadleaves. As shown, although present in
two of three sites, Eucalyptus globulus has the largest car-
bon accumulation estimated at 43% of the total amount. It
is far followed by Cupressus lusitanica. In addition, other
species, even though widespread in all the three sites do not
offer significant carbon stocks, especially Podocar pus u s a-
mbarensis and Jacaranda mimosifolia (Fig. 3).
Discussion
Limitations and uses
Study sites were intentionally selected to belong to the in-
stitutional domain. Knowing that urban area comprises in-
stitutional, residential as well as street trees, data on tree
density as well as carbon distribution in this study were not
extrapolated to the entire city. Thus, with due caution, cur-
rent results should be considered at the institutional scale
while expecting that future studies have to expand the sam-
ple size to ensure generalizable conclusions.
Urban species composition
Based on the three study sites, the city of Bukavu is made
of both native and exotic tree species. The presence of four
species of conifers and other broadleaves such as Eucalyp-
tus globulus from temperate zones explains the higher flo-
ristic diversity of the city. Specifically, conifers found are
probably the work of White Catholic missionaries who
managed all the three selected sites in Bukavu. Indeed,
these forested lands rather result from artificial afforestation
by institutional initiatives. Thus, this tree composition con-
firms (Nowak et al. 2007; Gulsvig et al. 2010) that the ur-
ban vegetation is a mix of native and non-native species in-
troduced by residents or other means making a tree diver-
sity often higher than in the surrounding native landscapes.
As for species nature, results reveal few fruit trees within
institutional urban sites of Bukavu. In fact, of 25 species
surveyed, only three are fruit trees, i.e. Mangifera indica,
Persea americana and Vernonia amygdalina which, more-
over, are not prevalent. This situation contrasts with that of

Urban Forest Cover and Carbon Dioxide Storage
88 Journal of Forest and Environmental Science http://jofs.or.kr
the residential area of Kinshasa which is mainly populated
by fruit trees. Indeed, Pauwels (1993) indicates that old cit-
ies are mainly planted to Mangifera indica, Persea americana,
Dacryodes edulis, Elaeis guineensis and Cocos nucifera.
Moreover, results show around 12% of ramified trees in
the city of Bukavu. The presence of abundant branched
stems in urban areas should be a consequence of reduced
competition for light as a result of their sparse distribution
pattern on ground. In fact, McHale (2009) and Bruyat
(2011) explain that low tree density in urban environments
reduce competition for light and other resources surround-
ing trees. Thus trees, instead of producing a tapered trunk
fetching light, tend to be wide by spreading their canopies.
More likely however, this could also be a result of free
growth due to poor tree management practice in these areas.
Diameter distribution and tree density
The pattern of tree distribution in diameter revealed a
relatively young population. In fact, trees with a DBH less
than 30 cm accounted for 62% of all trees surveyed in the
study sites. This size distribution differs from that of the
Oakland city where Nowak (1993) found approximately
61% of trees having less than 15 cm of DBH. On that,
Nowak (1994b) explains that the distribution of tree sizes
in urban area usually varies depending on the history and
intensity of vegetation management.
Irrespective of sites, the density averages 54 trees per ha
in Bukavu. This value is far less than estimates reported for
african rainforest ranging from 300 to 700 trees per ha
(Dupuy et al. 1998). It is still very low compared to the
density of institutional lands in the Oakland city which
reaches 111.9 trees per ha (Nowak 1993). Therefore, low
presence of trees in the urban landscape of Bukavu is note-
worthy, and then worrisome given the current climate
change challenge.
As for the basal area, while Dupuy et al. (1998) report
values between 30 and 35 m2 ha-1 in the african rainforest,
the average in the city of Bukavu was 5.21 m2 ha-1. This low
basal area is explained by the extent of infrastructures and
other dedicated spaces (streets, roads, sidewalks, lawns,
playgrounds...) dominating in urban environment.
With regard to density, Acacia spp, Cedrela serrata and
Markhamia lutea, although being among the most preva-
lent species in stems per ha do not cover necessary the larg-
est basal area. This is due to their smaller girth at DBH de-
spite their abundance as compared to some others like
Albizia gummifera, Melia azedarach and Podocarpus usa-
mbarensis. Therefore, to identify species with the greatest
diameter among urban trees of the same age can lead the
choice for efficient species to plant in the city.
Assessing Carbon dioxide storage variation in Bukavu
Total carbon stored in the aboveground biomass of trees
differed from one site to another. Thus 58.7 tons, 107.4
tons and 146.7 tons of carbon were recorded respectively in
Cathédrale Notre-Dame de la Paix, Collège Alfajiri and
Institut Supérieur Pédagogique. The average of carbon stored
in these areas amounted to 15.1 tons ha-1. This value is
nearly 3.6-fold lesser than that reported by Imani et al.
(2016) in the natural forest of the Kahuzi-Biega National
Park (54 tons C ha-1) located nearby the city of Bukavu.
Thus it reveals the low tree density of urban forests as com-
pared to that of natural forest vegetation outside urban area.
The difference observed in carbon stock average could also
be a consequence of the dissimilarity in species composition
(Fayolle et al. 2013) between these artificial and natural
ecosystems.
Results showed also that carbon density differs among
sites with S-ND being significantly higher than S-CA (p＜
0.05) (data not shown). This is essentially due to differ-
ences in current tree density in the three sites of Bukavu.
Therefore, identifying tree optimal number per area as well
as the nature of trees to plant in respect to their own charac-
teristics, including size at maturity and carbon storage po-
tential in any city is worthy of research.
Compared capacity of species in carbon storage
Most tree species holding the greatest carbon stock pre-
vailed also in both density and basal area. These are mainly
Eucalyptus globus, Cupressus lusitanica, Grevillea robusta
and Jacaranda mimosifolia (Table 4). For some other spe-
cies, however, relatively high density and or higher basal
area did not account for greater carbon storage. Indeed, this
is due to many factors such as species, age, size at maturity
and tree growth rate which greatly influence carbon seques-
tration and storage (McPherson 1998; Nowak et al. 2002;
Nowak and Crane 2002; Guarna 2012).
Moreover, tree diameter and moisture content do directly

Bakach D. KADIATA and
J
. B. Ncutirakiza NDAMIYEH
E
J For Environ Sci 33(2), 79-90 89
affect biomass calculation. Yet, conifers generally have low-
er moisture content than hardwood, 0.48 against 0.56 on
average (Nowak 1994a). Thus, it is not surprising that a
hardwood contains more carbon than a conifer of the same
or lesser diameter girth. Hence, beside their abundance,
hardwoods store nearly 6-fold much more carbon in
Bukavu (268.4 tons C against 44.3 tons C for conifers) due
to their own characteristics including their large size at
maturity. This is also the reason why the average of carbon
stored per ha under the tropics remains higher than that of
woodlands dominated by conifers.
As regards species, Eucalyptus globulus particularly
stores more carbon than any of the other species under
study in Bukavu, making nearly half of total carbon stock of
selected urban lands. This should be due to its great adapt-
ability to the area as well as to its high growth rate beside its
abundance in the landscape.
Environmental role of trees in institutional lands of
Bukavu
According to Vergiette and Labrecque (2007), an automo-
bile rejects about 4,500 kg of carbon after running 20,000 km.
This ratio is equivalent, all other things being equal, to 225 kg
of carbon released after 1,000 km of distance. The total car-
bon stored in the three institutional sites in Bukavu amou-
nting to 312.8 tons, represents emissions of nearly 1,390 ve-
hicles having traveled 1,000 km. This environmental serv-
ice is not, in any case, negligible during this century given
the global warming phenomenon already threatening hu-
man life and many natural ecosystems survival. No doubt,
there is great need to increase the tree component in urban
environments.
Conclusion
Assessing the ability of urban trees to reduce atmos-
pheric gas emissions including CO2 remains worthy of re-
search given the current environmental challenge.
Based on the diameter at breast height of 1,113 trees in-
ventoried, the potential of carbon storage varies from one
species to another within each site, and among institutional
sites in the city of Bukavu. Thus, 58.7 tons C, 107.4 tons C
and 146.7 tons C are respectively stored in Cathédrale
Notre-Dame de la Paix, Collège Alfajiri and Institut Supérieur
Pédagogique of Bukavu. Of 312.8 tons of carbon stored by
the three sites amounting to 1,147.9 tons of carbon dioxide
equivalents, broadleaves store approximately 268.5 tons C
representing 85.8% of the total storage while conifers store
only 44.3 tons C. The greatest carbon-fixing tree species is
Eucalyptus globulus storing about 134 tons C or 43% of total
stock, followed by far by Cupressus lusitanica. The average of
carbon stored per hectare reaches 15.1 tons.
Thus, trees surveyed in the urban area of Bukavu play a
significant role in reducing atmospheric CO2 and, thereby
contributing to mitigate global warming effects. Given the
imperative concern to combat climate change, more studies
are needed to enrich tree database in Bukavu covering both
institutional and residential lands as well as street trees.
This need remains also crucial for other urban areas in
DRC towards good management of urban forest in order to
improve environmental quality and citizens’ comfort. Pro-
grams to increase the potential for carbon sinks in urban
areas are therefore required.
Acknowledgements
We sincerely thank the FOGRN-BC project and the
German Cooperation Agency (GIZ) for the logistical sup-
port provided for this research. Thanks are due to the
Faculty of Agricultural Sciences of the University of Kin-
shasa for data collection equipment supplied. We are grate-
ful to the botanists Franklin BULONVU and Ghilain
ANDANGA, for their contribution to identifying species
on the field. At last, our gratitude goes to the responsibles of
Cathédrale Notre-Dame de la Paix, Collège Alfajiri and
Institut Supérieur Pédagogique of Bukavu for having al-
lowed the undertaking of this study within their premises.
References
Aguaron E, McPherson EG. 2012. Comparison of Methods for
Estimating Carbon Dioxide Storage by Sacramento’s Urban
Forest. In: Carbon Sequestration in Urban Ecosystems (Lal R,
Augustin BJ, eds). Springer Science＋Business Media, Dord-
recht, Netherlands, pp 43-71.
Anonymous. 2005. Monographie de la province du Sud-Kivu.
Unité de pilotage du processus DSCRP, Kinshasa, Gombe, 124 pp.
Anonymous. 2012. Région du Lac Kivu et d’autres de Grands
Lacs. Agence espérance. Accessed 11/27/2013.

Urban Forest Cover and Carbon Dioxide Storage
90 Journal of Forest and Environmental Science http://jofs.or.kr
Bell R, Wheeler J. 2006. Talking Trees: An Urban Forestry Toolkit
for Local Governments. Local Governments for Sustainability,
U.S., 88 pp.
Brack CL. 2002. Pollution mitigation and carbon sequestration by
an urban forest. Environ Pollut 116 Suppl 1: S195-S200.
Bruyat G. 2011. Estimation de la masse du carbone stockée par les
arbres de la communauté urbaine de Lyon. Edition du Certu,
Lyon, France, 88 pp.
Center for Urban Forest Research. 2008. CUFR Tree Carbon
Calculator - Help Menu and Background. United States Forest
Service, U.S., 34 pp.
Dupuy B, Durrieu de ML, Petrucci Y. 1998. Sylviculture des
peuplements naturels en forêt dense humide africaine : acquis et
recommandations. Bois et Forêts des Tropiques 257: 5-22.
Fayolle A, Doucet JL, Gillet JF, Bourland N, Lejeune P. 2013.
Tree allometry in Central Africa: Testing the validity of pan-
tropical multi-species allometric equations for estimating bio-
mass and carbon stocks. For Ecol Manage 305: 29-37.
Guarna B. 2012. An Analysis of Carbon Sequestration on Clarkson
University’s Campus. Capstone, 36 pp.
Gulsvig A, Frieder A, Wilson K, Potter S, Hertel W. 2010.
Residential Tree Inventory and Assessment. Environmental
Science, Policy and Management 4041w, St Louis Park, 51 pp.
Imani MG, Zapfack L, Mangambu M, Penedimanja P, Mwanga
MI, Boyemba F, Amani C. 2016. Caractérisations structurale,
floristique et biomasse ligneuse du Peuplement à Hagenia abys-
sinica du Parc National de Kahuzi-Biega (Rift Albertin, RDC).
European Scientific Journal 12: 189-209.
Lessard G, Boulfroy E. 2008. Les rôles de l’arbre en ville. Centre
collégial de transfert de technologie en foresterie de Sainte-Foy
(CERFO), Québec, 21 pp.
McHale MR, Burke IC, Lefsky MA, Peper PJ, McPherson EG.
2009. Urban forest biomass estimates: is it important to use allo-
metric relationships developed specifically for urban trees? Urban
Ecosystems 12: 95-113.
McPherson EG. 1998. Atmospheric carbon dioxide reduction by
sacramento’s urban forest. Journal of Arboriculture 24: 215-223.
Nowak DJ, Crane DE, Stevens JC, Hoehn RE, Walton JT, Bond
J. 2008. A ground-based method of assessing urban forest struc-
ture and ecosystem services. Arboriculture Urban Forestry 34:
347-358.
Nowak DJ, Crane DE. 2002. Carbon storage and sequestration by
urban trees in the USA. Environmental Pollution 116: 381-389.
Nowak DJ, Hoehn RE, Crane DE, Stevens JC, Walton JT. 2007.
Assessing Urban Forest Effects and Values: Philadelphia’s
Urban Forest. Resource Bulletin NRS-7. U.S. Department of
Agriculture, Forest Service, Northern Research Station, New-
town Square, PA, 22 pp.
Nowak DJ, Stevens JC, Sisinni SM, Luley CJ. 2002. Effects of ur-
ban tree management and species selection on atmospheric car-
bon dioxide. Journal of Arboriculture 28: 113-122.
Nowak DJ. 1993. Atmospheric carbon reduction by urban trees.
Journal of Environmental Management 37: 207-217.
Nowak DJ. 1994a. Atmospheric Carbon Dioxide Reduction by
Chicago’s urban forest. In: Chicago’s urban forest Ecosystem:
Results of the Chicago Urban For e s t C l i m at e Pr o j e c t ( M c -
Pherson GE, Nowak DJ, Rowntree RA, eds). United States
Department of Agriculture, Forest Service, Northeastern Re-
search Station, U.S., pp 83-94.
Nowak DJ. 1994b. Understanding the structure of urban forest.
Journal of Forestry 92: 42-46.
Pauwels L. 1993. Nzayilu N’ti. Guide des arbres et arbustes de la
région de Kinshasa-Brazzaville. Volume 4. Jardin botanique na-
tional de Belgique et Commission des Communautés Europée-
nnes. Scripta Botanica Belgica, Bruxelles, 495 pp.
Peck SW, Callaghan C, Monica EK, Bass B. 1999. Greenbacks
from Green roofs: Forging a new Industry in Canada. Environ-
mental Adaptation Research Group, Environment Canada, 78 pp.
Polepole P. 2007. Evacuation des déchets solides industriels dans la
ville de Bukavu. Cas de la Pharmakina et de la Bralima. ISP
Bukavu, Mémoire online, Bukavu, RDC, 52 pp.
Vergriete Y, Labrecque M. 2007. Rôles des arbres et des plantes
grimpantes en milieu urbain : revue de la littérature et tentative
d’extrapolation au contexte montréalais. Institut de recherche en
biologie végétale - Jardin Botanique de Montréal. Université de
Montréal, Canada, 35 pp.
Yang J, McBride J, Zhou J, Sun Z. 2005. The urban forest in
Beijing and its role in air pollution reduction. Urban Forestry
and Urban Greening 3: 65-78.