BAHIR DAR UNIVERSITY

GRADUATED PROGRAM IN ENVIRONMENT AND CLIMATE

CHANGE DEVELOPING SITE SPECIFIC ALLOMETRIC MODELLS TO ESTIMATE
ABOVEGROUND BIOMASS OF EUCALYPTUS GLOBULUS OF SMALLHOLDER
PLANTATION IN WOGERA DISTRICT, CENTRAL GONDAR HIGHLANDS OF
ETHIOPIA

Msc Thesis
BY;
Fentanye Birhan
SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF MASTER OF SCIENCE (MSc) IN ENVIRONMENT AND CLIMATE
CHANGE PROGRAM

May, 2024
Bahirdar Ethiopia
 BAHIR DAR UNIVERSITY

GRADUATED PROGRAM IN ENVIRONMENT AND CLIMATE CHANGE

DEVELOPING SITE SPECIFIC ALLOMETRIC MODELLS TO ESTIMATE
ABOVEGROUND BIOMASS OF EUCALYPTUS GLOBULUS OF SMALLHOLDER
PLANTATION IN WOGERA DISTRICT, CENTRAL GONDAR HIGHLANDS OF
ETHIOPIA
MSC THESIS
BY;
FENTAYE BIRHAN
SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF MASTER OF SCIENCE (MSc) IN ENVIRONMENT AND CLIMATE
CHANGE PROGRAM

Major Advisor: Beyene Belay (PhD)

Co – Advisor: Yohannis Gebremariam (PhD candidate)

May, 2024
Bahirdar Ethiopia
 THESIS APPROVAL SHEET

As members of the Board of Examiners of the Master of Sciences (M.Sc.) thesis open defense
examination, we have read and evaluated this thesis prepared by Fentaye Birhan entitled
DEVELOPING SITE-SPECIFIC ALLOMETRIC MODEL TO ESTIMATE THE
ABOVEGROUND BIOMASS OF THE Eucalyptus globulus OF SMALLHOLDER
PLANTATIONS IN WOGERA DISTRICT, HIGHLANDS OF CENTRAL GONDAR.
We here certify that the thesis is accepted for fulfilling the requirements for the award of the
degree of Master of Sciences (M.Sc.) in Environment and Climate Change.

Board of Examiners

____________________ ____________________ ___________________

Name of the external examiner signature date

____________________ ____________________ ___________________

Name of the internal examiner signature date

____________________ ____________________ ___________________

Name of the chairman signature date

____________________ ____________________ ___________________

DECLARATION

This is to certify that this thesis entitled “DEVELOPING SITE-SPECIFIC ALLOMETRIC

MODELS TO ESTIMATE ABOVEGROUND BIOMASS OF Eucalyptus globulus OF
SMALLHOLDER PLANTATIONS IN WOGERA DISTRICT, HIGHLANDS OF

i
CENTRAL GONDAR” submitted in partial fulfillment of the requirements for the award of
the degree of Master of Science in “Environment and Climate Change” to the Graduate
Program of College of Agriculture and Environmental Sciences, Bahirdar University by Miss.
Fentaye Birhan (ID. No. BDU1300998) is an authentic work carried out by her under my
guidance. The matter embodied in this project work has not been submitted earlier for the
award of any degree or diploma to the best of my knowledge and belief.

Name of the Student

Fentanye Birhan

Signature & date

28/5 /2024

Name of the Supervisors

Beyene Belay (PhD) (Major Advisor)

Signature & date

28/5/2024

Yohannis Gebremariam (PhD candidate) (Co Advisor)

Signature & date

28/ 5 2024

AKNOLOWEGEMENT

My first gratitude is to the Almighty God. Who gave me health and strength during my study
and being with me everywhere. Next, I would like to extend my great and deepest thanks to

ii
my advisor Beyene Belay (PhD). His advice, valuable comments, suggestions and corrections,
supporting any time from the beginning up to the finalization of my thesis. Besides, my
deepest gratitude to my co-advisor, Yohanis Gebremariyam (PhD candidate) for his critical
comments and kindness in supporting me. Furthermore, I would like to thank my instructor
Asmamaw Alemu (PhD) for his demonstrative appreciation, reinforcement and comments. I
need and I greatly acknowledge Bahirdar University for giving me a chance BDU Talent
scholarship for preparing this study. I am also happy to acknowledge Gondar University and
particularly those in the department of Forestry for allowing all research facilities that were
used during my study and my staff’s M/r Amare Gebeye and Kirobel Motebayenore helping
during data collection all fieldwork and giving me comments. Last but not least, this study
would have never been what it is now, had it been not for the sponsorship of the Restoration of
Sacred Forests and Degraded Landscapes for Climate Resilience and Livelihood Improvement
project for their financial support at the end I acknowledge my colleagues of the BSc teachers.

DEDICATION

To my sister Alemaz Motbaynore, great inspirers of my heart and mind. She does not live in
Ethiopia, but she is always with me. I am so proud of you.

iii
 ABBREVIATION/NOMENCLATURE

AGB Aboveground biomass......................................................................................1

AIC Akaike Information Criterion..........................................................................21
C Carbon................................................................................................................9
CO2 Carbondioxide...................................................................................................8
D Diameter............................................................................................................3
DBH Diameter at breast height...................................................................................2

iv
GHGs Green House gas...............................................................................................8
H Total tree height................................................................................................2
MAPE mean absolute percentage error.....................................................................23
REDD+ Reducing Emissions from Deforestation and Forest Degradation...................2
RMSE) Root mean square error................................................................................23
RSE Residual Standard Error................................................................................21
V volume...........................................................................................................19
ρ Wood density...........................................................................................................................19

ABSTRACT

In Ethiopia, smallholder farmers commonly grow eucalyptus plantations in their farmland and
other communal areas for short-rotation harvests (4–5years).Its expansion and socioeconomic
value of Eucalyptus plantation have been studied in the study area. Hence, this study was
conducted to develop a site-specific allometric model to estimate aboveground biomass and
height predictions for this smallholder plantation using the destructive method. The sample
design was applied using both purposive and simple random sampling techniques. A total of
60 trees with diameters at breast height ranging from 2.1 to 12.6 cm were randomly harvested
and separated into tree components (stem, branch, leaf, and twigs) weighted in the field and in

v
the laboratory for sample dry and fresh weight analysis. Model development was done using
75% of the total data set, whereas the remaining 25% was used for model validation. Eight
different aboveground equations and three height prediction equations were developed and
tested. Diameter at the breast height, total height, and woody density were used as predictors,
and aboveground biomass was used as a dependent variable. For the height prediction model,
the diameter was used as an independent variable. The best aboveground biomass models
were selected based on the highest coefficient determination, lowest values of residual
standard error, Akaike information criteria, and coefficient factor, and validated using the
mean absolute percentage error and root mean square error. Site-specific models using the
three simple predictor variables fitted the data in the form of: ln(Y) =-2.875+ 1.243ln (D) +
1.631ln (H)+ 0.665ln (p), both of which met good fitness criteria and were validated. This
model has coefficient determination (adjiR 2) (0.94), residual standard error (0.230), Akaike
Information Criteria (1.297), a correction factor (1.027), a root mean square error(2.88) and
a mean absolute percentage error (15.71%). Comparing the current model (M6) with the four
general existing models, the current model approximately predicted the smallholder
plantation better than the general existing model, followed by model lnY= -2.187+ 0.916
ln(D^2H ρ) / 0.112 (D^2H ρ) ^0.916 The best-fit model for the height prediction model was
model (M3) with adjiR2 (0.84), RSE (0.102), and AIC (-78) in the form lnH = 1.400 + 0.5
(lnD). Based on the findings, it is possible to conclude that the developing site-specific models
are preferable to accurately estimate aboveground biomass and height compared to the
existing or previously developed general models. Key words; Aboveground biomass,
Allometric model, Eucalyptus globulus and tree height.

TABLE OF CONTENTS

CONTENTS pages

THESIS APPROVAL SHEET.....................................................................................................i

DECLARATION.........................................................................................................................ii

AKNOLOWEGEMENT............................................................................................................iii

vi
DEDICATION............................................................................................................................iv

ABBREVIATION/NOMENCLATURE.....................................................................................v

ABSTRACT...............................................................................................................................vi

TABLE OF CONTENTS..........................................................................................................vii

CONTENTS..............................................................................................................................vii

LIST OF TABLES.......................................................................................................................x

LIST OF FIGURES....................................................................................................................xi

LIST OF APPENDIX FIGURES..............................................................................................xii

Chapter 1. INTRODUCTION.....................................................................................................1

1.1. Background.......................................................................................................................1

1.2. Statement of the Problem..................................................................................................3

1.3. Objectives..........................................................................................................................5

1.3.1. General Objective.......................................................................................................5

1.3.2. Specific Objective......................................................................................................5

1.4. Research Questions.......................................................................................................5

(Continued)

Chapter 2. LETRURTURE REVIEW.........................................................................................6

2.1. Eucalyptus Plantation in Ethiopia.....................................................................................6

2.2. Role of Smallholder Plantation.........................................................................................7

2.3 Aboveground Biomass Estimation.....................................................................................7

vii
 2.4. Aboveground Carbon stock Estimation............................................................................8

2.5. Allometric Models..........................................................................................................10

2.6. Height Prediction Model.................................................................................................12

Chapter 3. MATERIALS AND METHODS.............................................................................14

3.1. Description of the Study Area.........................................................................................14

3.1.1. Climate and Soil Type..............................................................................................14

3.1.2. Land Use Type of the Study Area............................................................................15

3.2. Sampling Design and Techniques...................................................................................15

3.2.1. Biomass Measurement.............................................................................................16

3.2.2. Sample tree Selection...............................................................................................16

3.2.3. Aboveground Biomass Estimation...........................................................................16

3.3. Laboratory Measurement................................................................................................17

3.4. Data Analysis..................................................................................................................18

3.5. Allometric Equation Development.................................................................................19

3.5.1. Model Development................................................................................................19

(Continued)

3.5.2. Model Selection and Validation..............................................................................20

3.6. Comparison of the Current Model with the previous existing General Model..............22

3.7. Tree Height Prediction Model Development and Selection...........................................23

Chapter 4. RESULT AND DISCUSSION................................................................................24

viii
 4.1. Characteristics of the Measured Trees............................................................................24

4.2. Model development and selection...................................................................................25

4.3 Model Validation.............................................................................................................30

4.4. Comparison of the Current Model with the Existing General Models......................32

4.5. Height Prediction Model Development and Selection................................................34

Chapter 5. CONCLUSION........................................................................................................37

Chapter 6. RECOMMENDATION...........................................................................................38

REFFERENCE..........................................................................................................................39

APPENDIX................................................................................................................................55

BILOGRAPHY..........................................................................................................................57

LIST OF TABLES

Table.......................................................................................................................................page

Table 1:Allometric equation types tested for estimating total aboveground biomass For
E.globule s.................................................................................................................................19

Table 2:selected previously published model s.........................................................................22

ix
Table 3: Allometric equation types Developed for total height prediction model for E.globules
...................................................................................................................................................23

Table 4:Statistical summary for tree parameters used in developing or fitting and validating
allometric models for aboveground biomass estimation and height prediction model............24

Table 5:Developed allometric model to estimate total aboveground biomass..........................29

Table 6: Mean absolute percentage error (MAPE(%) and relative root mean square error
(RSME) for AGB estimation allometric equations developed using a 25% (17) independent
set of data...................................................................................................................................30

Table 7: comparison of the current best selected model with the existing models using the
whole measured data (N=60).....................................................................................................32

Table 8: Developed Allometric model to prediction total height of E.globuls smallholder

plantation...................................................................................................................................35

LIST OF FIGURES

Figure ....................................................................................................................................page

Figure 1: Study Area Map.........................................................................................................14

Figure 2:Scatter plot of tree aboveground tree biomass (AGB) vs. potential predictor variables
(diameter, height, and wood-specific density)...........................................................................25

x
Figure 3: Scatter plots of residual values versus predicted value of the three best allometric fit
models Predicted value in log scale...........................................................................................31

Figure 4: Comparison of the current model with previously general equation used to estimate
aboveground biomass of smallholder plantation of Eucalyptus globulus tree..........................34

Figure 5: scatter plots of Total height versus the potential of predictor Diameter....................35

Figure 6: residual and observed plots against predicted height for the best-fit model for this
study...........................................................................................................................................36

LIST OF APPENDIX FIGURES

Appendix Figure.....................................................................................................................page

Appendix 1: Figure 1: Photo taken during field data collection...............................................55

Appendix 1: Figure 2:The photos show laboratory analysis of samples taken in the Laboratory
...................................................................................................................................................56

xi
xii
 Chapter 1. INTRODUCTION

1.1. Background and Justifcation

Eucalyptus species are well-adapted and familiar tree species in various weather conditions
around the world. About 20 million hectares are planted with them, mostly in temperate,
tropical, and subtropical climates. These plants are among the globally most farmed forest
species (Tsegaye Bekele, 2015; Mekonnen Daba, 2016). The word Eucalyptus is derived from
the Greek words ‘eu’ and 'kalyptos', which are exotic to Ethiopia and are currently widely
planted for various uses in different parts of the country. It belongs to the family Myrtaceae,
an ever-green flowering tree that consists of some 800 species. It is the dominant tree species
commonly found in the central highlands of Ethiopia since 1895 ( Mesfin Abebe &Wubalem
Tadesse, 2003; Haileab Zegeye, 2010).

smallholder Ethiopian farmers frequently plant eucalypt tree species on their farmlands. As a
result, since the 19th century, eucalyptus plantations in the shape of woodlots have greatly
expanded throughout the Amhara region, including the study area. Currently, even smallholder
farmers are shifting their crop land to Eucalyptus plantations (Yordanos sete & Getahun
Abreham, 2021). Additionally, as eucalyptus plantations become more popular as a source of
charcoal and fuel wood for humans, they reduce pressure on remaining natural forests. These
planation forests not only produce wood products but also contribute significantly to the
mitigation of climate change by sequestering and fixing carbon (Hartono & Kusmana, 2020).
which can provide economic, environmental, and social benefits in addition to helping to
prevent climate change through carbon sequestration.

In tropical areas, cutting carbon emissions from deforestation and degradation is a major
opportunity (Canadell & Raupach, 2008). This study focused on aboveground biomass (AGB)
because it represents the largest carbon pool of forest vegetation and is directly impacted by
deforestation and degradation (Kenzo et al., 2009; Basuki et al., 2022).
The primary use of allometric regression models is as mathematical functions that connect a
tree's dry mass to one or more tree dimensions; diameter at breast height (DBH) and total tree
height (H) are the two most important single tree metrics that describe the composition and
structure of forests (H) and also include woody density (Ebuy et al., 2011; Sturrock et al.,
2011 Birhanu Kebede et al., 2018).

Most of the allometric models now in use for estimating tree aboveground biomass in the
tropics were created from forest ecosystems (Brown et al., 1989; Brown et al., 1997; Chave et
al.,2005; 2014). However, due to their inherent uncertainty, these models are unable to
sufficiently reveal the tree biomass in a given region or limited area in the tropics, including
Ethiopia. To attain greater degrees of accuracy and precision, it is crucial to apply site-specific
and species-specific allometric equations (Djomo et al., 2010; 2016; Admassu et al., 2022).
This is because different tree species have different allometric biomass estimations, and also
because the wood density of the species, the floristic composition, and the growth
performance of the species-environment affect the type of allometric model (Kaonga Martin &
Smith Bayliss, 2010).

Therefore, site-specific allometric equations are the most accurate and precise. They are
recommended for estimating the biomass of tree species. This is vital for monitoring,
reporting, and verification of components under the emissions trading schemes of Reducing
Emissions from Deforestation and Forest Degradation (REDD +) through sustainable forest
management mechanisms to attain credibility in the carbon markets (Gibbs et al., 2007;
Vieilledent et al., 2012). Developing a reliable way for measuring biomass carbon associated
with trees in smallholder plantations is critical for helping smallholder farmers benefit from
the emerging carbon market and benefit from carbon offset schemes (Roshetko,
2015;Pravalprukskul, 2018).

A site-specific allometric model that was constructed helps in the understanding of the
functions played by smallholder plantations in the global carbon cycle, especially climate
change. To study the carbon stock, site-specific allometric models are helpful and Accurate
biomass estimation is essential to measuring changes in both biomass and carbon stocks
(Banerjee et al., 2022). Plantations owned by smallholder farmers are not the same as
plantation forest.
These frequently supply smallholder plantations with fuel, charcoal, and poles so that the
smallholders may manage the plantations and ensure their sustainable use (Nambiar et al.,
2021). In the other sides When measuring height in an area where it is time-consuming,
unreliable, and more challenging to measure in a clothing canopy than the D; parameter, the
height with D relation model is one of the most crucial models (Mugasha et al., 2013; 2019).
This study also focused on developing An allometric model based on site was used to estimate
the aboveground biomass of Eucalyptus globulus smallholder plantation trees, taking into
account the harvesting method, robustness testing, and height prediction models. The
developed models reduce time, labor, error, and cost in the management of Eucalyptus
globulus smallholder plantations around the study area that are similar to the studied site and
give an accurate estimation of aboveground biomass.

1.2. Statement of the Problem

In the Ethiopian highlands, this species is among the most exotic and has great socioeconomic
and environmental significance. Eucalyptus woodlots are mostly established by Ethiopian
smallholder farmers as a source of income (Abebe Birara et al, 2019). For REDD+ as well as
climate adaptation and mitigation, accurate aboveground biomass estimation is essential. In
order to increase productivity and store a known quantity of carbon in biomass both above and
below ground, including eucalyptus species, forest plantations are a valuable resource for
forest inventories and biomass assessment (Ostadhashemi et al., 2014).

A generic allometric equation for carbon allocation and aboveground tree biomass estimation
in tropical and subtropical eucalypt plantations has been studied by many researchers for the
eucalyptus species (Williams et al., 2005; Gonzalez et al., 2013; Missanjo et al., 2015;
Gonzalez et al., 2021). However, general models create uncertainty, which yields over and
underestimated yields (Daba and Teshome Soromessa., 2019). Allometric models must be
developed based on site and species to get accurate and precise estimates. In most studies,
biomass and carbon quantities of trees vary with soil, management, climate, and species. For
enhanced biomass estimates of study sites, site-specific biomass models are needed (Mugasha
et al., 2012).
The AGB models depend on several variables, including tree height, which necessitates
accurate measurement. However, accurate tree height measurement in the field is labor-
intensive, costly, time-consuming, prone to measurement errors, and challenging to measure in
closed canopy settings, and tree heights vary by site and species, even if in the same species.
For this reason, D height relation models are empirically based on the D variable, which is the
most readily measured, least time-consuming, and capable of predicting individual tree height
(Shamaki et al., 2016).

In the study area Wogera district, most plantations are Eucalyptus globulus (Nich bahirzaf)
owned and managed by smallholder farmers. Various research are carried out in the study area
of smallholder plantation trees concerning their Expansion, role, effect, ecological and
Socioeconomic approach, and other elements. In several regions of Ethiopia (Tegegne Derbe
et al., 2018; Abebe Birara. et al., 2019; Yordanos sete & Getahun Abreham, 2021).In another
place ( Mulugeta Zewdie, 2008; Mulugeta Zewdie et al., 2009;Beleyenehe Dessie et al.,
2022).

However, there is still a research gap in the process of developing aboveground biomass
models. No scientific data was gathered and examined concerning site-specific allometric
models for estimating aboveground biomass and height prediction models, both locally and
nationally, to predict the aboveground biomass of the smallholder plantations in my study
area, we conducted this research. To accurately predict tree-level biomass in smallholder
plantations and the height model for Eucalyptus globulus.

In the field and laboratory, measurement was done directly using the above-ground biomass
harvesting method. 60 Eucalyptus globulus trees were chosen at random, and the predictor
variables (D cm), (H_m), and woody density g/cm-3 were taken into consideration for the ages
of 4 and 5. The developed site-specific allometric models for eucalyptus differ from existing
generalized allometric equations for tropical trees. Which can be used to produce accurate
estimates of aboveground biomass in an Eucalyptus globulus smallholder plantation tree with
a diameter range of 2.1–12.6 and a height range of 5–14 in the same geographical area.
 1.3. Objectives

1.3.1. General Objective

This study aimed to develop site-specific allometric models for estimating the aboveground
biomass of Eucalyptus globulus smallholder plantation in wogera district central Gondar High
lands of Ethiopia

1.3.2. Specific Objective

● To develop site-specific allometric models for estimating the aboveground biomass and

carbon stock for Eucalyptus globulus of smallholder Plantation

● To compare the current site-specific allometric modes with other general allometric

models

● To develop a height prediction model (Height-Diameter relationship) for Eucalyptus

globulus smallholder plantation

1.4. Research Questions

 What is an accurate site-specific allometric model for estimating the aboveground biomass
for Eucalyptus globulus smallholder plantation in the study area?
 How do we compare the current site-specific allometric models with other general
allometric models in the study area?
 What is an accurate site- specific height prediction model for Eucalyptus globulus
smallholder plantation in the study area?
 Chapter 2. LETRURTURE REVIEW

2.1. Eucalyptus Plantation in Ethiopia

Eucalyptus plantations have expanded considerably among smallholders and small forestry
firms in Ethiopia, placing them in second place in Africa in terms of area planted with
Eucalyptus ( Mesfin Abebe &Wubalem Tadesse, 2003;Biruk Ketsela, 2012).Eucalyptus trees
cover 90% of the total planted forest area in Ethiopia. This exotic species was first introduced
to Ethiopia around the end of the 19th century (Pohjonen & Pukkala,1990).Growing
eucalyptus began in Ethiopia around 1890, during the regime of Emperor Menelik II, to reduce
the shortage of wood. At the same time, Eucalyptus globulus was planted as the main tree
species in the central highlands of Ethiopia in the 19th century (Getahun Amare, 2003).

Eucalyptus is frequently planted in woodlots in my study area and is a great asset for planters
contributes to rural development and poverty reduction in Ethiopia (Abebe Birara et al., 2019).
For this tree, decisions about the species solely based on the perceived negative ecological
impacts are not sensible when it is planted on the right sites with good management planning
and proper tending operations focusing on its economic, social, and ecological impacts (Jagger
& Pender, 2003).

This tree also contributes to the contribution of climate change mitigation at the regional and
national levels, and its impact on the local community is highly imperative and essential for
mutual benefits (Alemayehu Asabeneh & Melka Yoseph, 2022). Trees also play a key role in
the global carbon cycle. Managing forests through agroforestry, forestry, and plantation
systems is seen as an important opportunity for climate change mitigation and adaptation
(Meragiaw Misganaw, 2017).

According to ( Ribeiro et al., 2015), plants, especially those with fast-growing species such as
Eucalyptus and its cultivars, represent a short-term and cost-efficient alternative for
sequestering carbon, which would then be emitted into the atmosphere. Eucalyptus species can
supply wood in good quantities within 4–5 years from relatively small areas, and this woodlot
plantation of land in the study area has been successfully established and increased since 1970
( Tegegne Derbe et al , 2018) .
 2.2. Role of Smallholder Plantation

A smallholder plantation is a plantation Managed forest, established for subsistence or fuel

wood, timber, fodder, and local sale by planting and seeding in the orchards, forest gardens
process of afforestation, with and other even age class and regular spacing (Herbohn and
Medrilzam, 2014; Obidzinski & Dermawan, 2010).

People in rural areas have long planted smallholder trees all across the tropics, including
Ethiopia (Fauzi & Rohadi, 2015). Smallholder trees These serve rural communities in a
variety of ways, providing fuel wood, fodder, and poles, building materials for daily use and
construction, as well as environmental benefits like helping to mitigate global and national
climate change and balancing greenhouse gas emissions that promote photosynthesis. (Nawir
et al., 2007). In my study area for Eucalyptus globulus smallholder tree has been studied by
many researchers but not all address, so this study conduct add the reliable estimation of
aboveground biomass site specific allometric model and height prediction model for biomass
estimation carbon stock and to climate change mitigation role through the correct estimation
of biomass.

2.3 Aboveground Biomass Estimation

The biomass of a forest is the total mass of living things there. It is the weight of all living
things in the environment, and it can be exactly calculated by weighing every living item.
(Condit, 2008). To model bioenergy, food security, environmental assessment, and climate
change research, biomass is a crucial element. Nevertheless, estimating biomass remains a
difficult undertaking, particularly in regions with complicated forest stand patterns and
environmental conditions (Lali et al., 2015). Most of the biomass in a forest is from trees, and
the focus of methods for estimating biomass is measuring the above-ground portion of trees
because it is directly impacted by deforestation and degradation and represents a large carbon
pool (Kanninen, 2005).
On the other hand, the part of the tree's biomass that is below the roots is much more difficult
to measure. For this reason, nearly all inventories of forest biomass refer only to aboveground
weight (Condit, 2008). Biomass is also an indicator of site productivity, both in biological and
economic terms, and it is important to maintain the implementation of sustainable forest
management and plans (Návar, 2009; Zianis & Mencuccini, 2005).

In this study, the fastest-growing and most successful smallholder plantations have
significance for socioeconomic purposes. Eucalyptus globulus smallholder tree developed a
site-specific allometric model and a height prediction model that were used to estimate the
above-ground biomass and carbon stock. Biomass is a function of diameter at breast height
(D), total height (H), and wood density in a given location and site. However, the contribution
of these variables to the above-ground biomass differs from site to site depending on the
succession stage of the forest, disturbance levels, species composition, age, management, etc
(Ebuy et al., 2011; Henry et al., 2011).

It is of great significance for determining the most effective way to analyze carbon storage,
dynamics, monitoring, and estimating lost or emitted carbon during deforestation, and it gives
us an idea of the forest’s potential to sequester and store carbon in the forest ecosystem
(Vashum KT, 2012). In particular, for dominating and extensively distributed, quickly
growing, short-term, and commonly grown smallholder eucalyptus species, the creation of
site-specific, dependable allometric models is advised to increase the accuracy of such
estimations (Fayolle et al., 2013; Lin et al., 2017; Moussa et al., 2019). This study also
attempts to estimate biomass involving parameters such as DBH, diameter at the beast height,
total tree height, and woody density with different regression equations.

2.4. Aboveground Carbon stock Estimation

Climate change is agreed as a global threat to the environment and human wellbeing. Green
House gas GHGs are produced by deforestation, burning fossil fuels, and industrial CO2
emissions. As a result, it has been proposed that using forests for agroforestry, plantations, and
reforestation is one of the best land management strategies for reducing atmospheric CO2
levels through photosynthesis. Additionally, over 70% of all Soil organic carbon and over
80% of all terrestrial aboveground carbon C are stored in forest ecosystems (Binyam,
20014).Forest carbon evaluation has received more attention after the Paris Agreement .It is
necessary to provide realistic values of the carbon stored in the terrestrial biosphere (Lorenz &
Lal, 2010; Calfapietra et al., 2012).

With the increasing concentration of carbon dioxide in the Earth’s atmosphere as the result of
deforestation there is need to estimate biomass and carbon pools in tropical forests (Houghton,
2005). This is, particularly, essential in Africa where reliable biomass data is lacking,
Accurate estimates of biomass are important for the assessment of carbon stocks toward
understanding carbon variations in response to world climate changes, monitoring ecological
processes such as wood production and nutrient cycling and sustainable forest management
(Vieira et al., 2008;Rahman et al., 2021).

These days, forests and woodland ecosystems serve as significant carbon sinks, and protecting
them is one of the most viable ways to combat the growing global warming threat (Barbati et
al., 2007; Lorenz & Lal, 2010). When the amount of carbon dioxide in the atmosphere and its
impact on climate change worldwide keep rising.To offer information on the worldwide
carbon budget and to understand the global climate, modeling the aboveground biomass of
forest and woodland ecosystems is necessary (Ekoungoulou et al., 2018; Löf et al., 2019;
Henry et al., 2020).

Accurate estimates of carbon stocks depend to a greater degree on the availability and
adequacy of the allometric equations that are used to estimate tree biomass (Yehualashet
Belete et al., 2019). One of the biggest and most productive sources of stored carbon
worldwide is smallholder farmers in developing countries. However, in developing nations,
determining, monitoring, and quantifying the amount of carbon that trees on smallholder farms
are removing from the atmosphere continues to be extremely difficult (Kuyah et al., 2013) .

REDD+ (Reduced Emissions from Deforestation and Degradation, Plus Sustainable Forest
Management, Conservation is also a significant essential mechanism to address the issues of
climate change in developing countries. It needs accurate, standard and reliable information of
forest carbon stock and biomass production (Toirambe et al., 2013; Missanjo et al., 2015).
In Ethiopia there is rare site specific allometric model for smallholder plantation developing
site specific allometric equations is necessary to accurately quantify the biomass and carbon
storage of the smallholder plantation tree.

2.5. Allometric Models

The accuracy and applicability of allometric models have a significant impact on the
estimation of aboveground biomass (AGB). For the purpose of measuring biomass and carbon
storage in the terrestrial forest ecosystem, allometric equations are crucial(Amsalu Abich et
al., 2019).Biomass equations are a common and cost-effective method to estimate the biomass
of tree species present in plantations (Ostadhashemi et al., 2014).

The aboveground tree components are estimated using biomass estimation equations, also
referred to as allometric regression models. These equations relate the tree dry mass (obtained
by the direct method) to measurable variables and/or tree dimensions, such as diameter at
breast height (DBH, at 1.3 m from the ground), total tree height (H), woody density, and other
variables (Yehualashet Belete et al., 2019).

For the best-performing stem diameter selection algorithm, a range of 17–95 individuals were
needed to achieve biomass predictions with a standard deviation within 5% of the mean. This
indicates that allometric models must be based on sufficient data to ensure accurate and
precise predictions (Oxburgh et al., 2015). Nevertheless, funding and the goal of the research
also influence how many sample trees are chosen. Sometimes factors unrelated to prediction
accuracy, like time and cost, determine the sample size (Chave et al.,2005).

Regression equations were developed for the accumulation of biomass with age across all
productivity classes for each forest type. Consequently, much of the uncertainty associated
with the estimates of total carbon storage and the rates of carbon accumulation in the world's
forests can be attributed to uncertainties at the local or regional level (Author et al., 2012).
There are two ways to measure aboveground biomass in the field : the direct method of
harvesting and the indirect method that uses the allometric equations DBH and total height
( Ravindranath &Ostwald, 2008). In this study, the direct measurement of tree aboveground
biomass (AGB) involved felling a representative number of trees and measuring their fresh
parts (steam, branch, leaf, etc.) used for model development in the field and the oven-dry
process in the laboratory.This method is high-cost and time-consuming, but in terms of model
accuracy and validation, it is the best method, and when the trees are harvested for short
rotation, it is possible to use this method for model development (Blake et al., 2021).

The researchers developed site and species-specific allometric models for aboveground
biomass estimation for different trees in different areas, such as in Ethipia Mana Angetu
Forest for (Olea europaea L. subsp. Cuspidate by (Birhanu kebede &Teshome Soromessa,
2018) studied by (Amsalu Abich et al., 2019; 2021) for, Combretum-Terminalia woodlands
and dry deciduous woodland ecosystems in Amhara, Northwest Ethiopia. In dry lands of
Ethiopia for the invasive species Prosopis juliflora (Dessie Assefa et al., 2022), and (Mulugeta
Zewide et al., 2009), for Eucalyptus Globulus in high lands of Ethiopia etc.

have used these techniques to develop allometric models with various sampling techniques
and allometric model selection criteria, which include the three predictor variables and the two
predictor variables height and DBH and also others parameter.They have come to the
conclusion that, in order to estimate aboveground biomass more accurately than general
models, each species requires an allometric model that is specific to its site and species in the
Ethiopian forest.

when there is an allometric difference and one model is not appropriate for a particular species
or location. The most straightforward technique for estimating above-ground biomass and the
carbon stocks stored in the forest ecosystem is the harvest method. The harvesting method is
limited to a small area or small tree sample sizes. Nevertheless, it determines the biomass
accurately for a particular area; to validate and be accurate, the model was the best method,
and the trees had to be harvested and weighted (Jibrin & Abdulkadir, 2015; Kaonga Martin &
& Smith Bayliss, 2010) .

However, until today, no specific study has been reported regarding the development of site-
specific allometric models for the estimation of smallholder plantations of Eucalyptus
globulus trees. Therefore, the aim of the aim of this study was to develop a site-specific
allometric model using this method to estimate the above-ground biomass of smallholders.
Generalized allometry exists for tropical trees, but site- and species-specific models are more
accurate and reduce biomass estimation uncertainty at the local level (Litton & Kauffman,
2008 ; Bosques & Secundarios, 2019; Amsalu Abich et al.,2019,2021).

The primary causes of the variance in biomass amongst woods are environmental differences
(Rocha et al., 2020). The factors that affect the regression biomass models are soil properties,
climate conditions, etc (DeLucia et al., 2000 ; Antonio et al., 2007; Urban et al., 2013)
Allometric biomass models vary in species and in site (Montagu et al., 2005). Due to this, site-
specific allometric relationships are required for the conversion of forest inventory
measurements to regional-scale estimates of forest carbon sequestration (Sawadogo et al.,
2010).

This study also developed site-specific allometric models for estimating the aboveground
biomass of the smallholder plantation Eucalyptus globulus, which is found in the study area
and is managed and owed by smallholder farmers. In addition to these, to determine the
strength and significance of the best-fitting models, compare the developed models to the
existing general allometric models when they are developed in tropical or elsewhere and
validate this allometric equation by using different statistical indices that were applied for
comparing and evaluating the statically.

Carbon exists as carbon dioxide, which is one of the greenhouse gases (GHG) in the
atmosphere and has an enormous influence on the impact of climate change. Therefore, a
common method to quantify carbon stocks in forests is the use of allometric regression models
to convert forest inventory data to estimates of aboveground biomass (Chave et al., 2014;
Mohd Zaki et al., 2018).

2.6. Height Prediction Model

One of the key characteristics of forest stands is the correlation between tree height (H) and
diameter at breast height (D). Furthermore, H is a necessary variable in models of volume and
aboveground biomass. But measuring height takes longer than measuring D. Error factors for
height measurements include visual obstructions, rounded crown shapes, bending trees, and
sloped terrain (Mugasha et al., 2013, 2019).
Allometric models used in above-ground forest biomass estimation depend on tree height,
diameter,and wood density as predictive variables (Sullivan et al., 2018). While measurements
of stem diameter and wood density can be obtained easily, tree height measurements can be
time-consuming and difficult. This often results in the exclusion of tree height as a variable for
predicting the aboveground biomass of trees.Other studies have demonstrated that including
tree height can significantly improve aboveground biomass estimates. Applicable height-
diameter models that make the estimation of tree height easier are typically constructed in
order to guarantee that tree height is included as a predictor variable in the estimation of
aboveground biomass (Rutishauser et al., 2013). Because there are large differences in tree
height for a given stem diameter, forest type, and location, incorporating tree height can
greatly improve aboveground biomass estimations (Feldpausch et al., 2011).

Tree height and diameter correlations In this way, it lowers the costs related to forest inventory
by creating models to estimate the heights of the remaining trees in the stand (Paulo et al.,
2011). Even within the same forest stand, different tree species have different H and D
relationships. The relationship varies even within the same species, with variations in stand
densities, tree sizes, species compositions, stand ages, and site conditions. This impact is
frequently taken into account when building models with extra variables (Poorter et al., 2006).

As a result, it's crucial to develop and validate accurate models for predicting tree height based
on D. Additionally, accurate tree height estimations are critical for evaluating biomass above
ground and have a significant impact on models of forest growth that are based on ecological
and ecophysiological processes and guarantee sustainable management. Assessments of the
biomass of forest stands have also typically been taken into account (Zhang et al., 2014).
 Chapter 3. MATERIALS AND METHODS

3.1. Description of the Study Area

The study was carried out in the central Gondar Amhara Region's Wogera district, which is
located at 12°46'06.5"N and 37°37'26.5"E on a map (Figure 1). The district is about 1821 km2
in size. The distances are approximately 32 kilometers from Gondar city and 213 kilometers
northwest of Bahir Dar, the Amhara region's capital. The region is located 2812 meters above
sea level on average. An estimated 250,493 people called the district home, of which 122,635
were women and 127,858 were men. With a population density of 137 people per km2, 94%
of the population resides in rural regions. (Kassegne Asenake et al., 2021).

Figure 1: Study Area Map

3.1.1. Climate and Soil Type

The study area is classified as follows by the Ethiopian agro-ecological classification:

highlands (56%), semi-lowlands (26%), lowlands (14%), and 4% frost. The average annual
temperature is between 14 °C and 33°C. From May to September, there is an orographic
rainfall rate. The range of the annual average rainfall depends on the geographical type and is
between 1000 and 1600 mm. Furthermore,the predominant soil type in the region is
combisol,which is recognized as fertile soil.

3.1.2. Land Use Type of the Study Area

The major land uses in the study area include cultivated land, grazing land, and the plantation
of exotic species. In the district, the majority of planation forests are made of Eucalyptus
globulus tree species and are commonly owned and managed by smallholder farmers. Almost
70% of the plantations are Eucalyptus globulus (Kassegn Asnakew et al, 2021).

When the farmers were between four and six years of age, the local producers harvested the
stand. However, five years of age was the most common harvesting age. The main stem was
cut down and used to make firewood, charcoal, and construction poles. Smallholder farmers
harvest the tree between March and May. The spacing between trees was on average 0.5m x
0.5 m, which is the traditional planting spacing practiced by most of the local farmers in the
study area. Native tree species like waneza (Cordia africana) Gesho (Rhamuns prinoides)
woyera ( Olea africana) are barely seen in the landscape of the study area.

3.2. Sampling Design and Techniques

Both purposive and simple random sampling were employed to collect the data in this study.
In the first step, Wogera district was selected purposefully due to the potential availability of
Eucalyptus globulus smallholder plantations. In the second step, Kossoye Keble was selected
based on consultation with district agriculture office experts, as the kebele has better
expansion of smallholder plantations and road accessibility for data collection. In addition, the
selected kebele has better planation cover with different age ranges to get representative
sample plots. Farmers in Kossoye Kebele were willing to harvest sampled trees. Then, a
fundamental sample inventory of smallholder plantations carried out in this study collected
data on Eucalyptus globulus smallholder plantation trees.Eucalyptus globulus smallholder
plantation trees.
3.2.1. Biomass Measurement

3.2.2. Sample tree Selection

For this study, Eucalyptus globulus planation with the ages of 4 and 5 years was considered, as
there was not enough planation to sample ages above 5 years. For each age class, six plots
measuring 10 m by 10 m were sampled on different smallholders’ planation sites to cover the
variation in stand growth resulting from management techniques, soil fertility, and other
parameters. All trees in each sample plot were measured for DBH with the help of diameter
tape. In addition, the total height of representative trees was measured using a hypsometer.
However, a total of 60 sample trees (30 trees for each age class) were harvested for model
development. That means, as the number of sample plots was six for each age class, five trees
were randomly selected on each sample plot. To select five sample trees for harvesting, a
random sampling technique was used from the list of trees recorded for DBH measurement on
each sample plot. The total tree height of the 60 sample trees was measured using a tape meter
after they fell to minimize errors.

3.2.3. Aboveground Biomass Estimation

the selected trees was measured and marked. Then felled trees were measured for total height
using a linear tape meter. The trees felled for biomass estimation were separated into stems,
brunches, leaves, and twigs. The fresh mass of each component was weighted in the field
using a 50 kg scale hanging balance with a precision of 0.1g. Before weighing the fresh
weight, the stems were cut into sections to make them suitable for hanging balance
measurements. Fresh weights of about 200 g for leaves, twigs, and branches were randomly
collected and placed in plastic bags.
Then samples in the plastic bag were tagged and taken to the laboratory for dry biomass
determination. Similarly, the stem disks, 5 cm thick, were sampled from different sections of
each sampled tree (base, middle, and top). In total, 180 stem disks were taken to the laboratory
after weighing the fresh weight for corresponding oven-dried biomass estimation. For wood
density estimation, many sub-sections 1 m to 2 m long were taken and immediately weighted
in the field. In addition, the diameter of each sampled log was measured both at the bottom
and top. Generally, the field measurement procedure and the biomass calculation in this
research followed the procedure of the FAO Food Manual for the destructive sampling method
prepared (Picard, 2012).

3.3. Laboratory Measurement

All samples collected in the field were taken to the laboratory at the University of Gondar,
Agricultural and Environmental Science College, for oven-dried biomass estimation. The
leaves were oven-dried at 70°C by monitoring changes until we reached a constant weight.
Stem desks and brunches were oven-oven dried at 105°C until all the moisture became zero or
constant weight (Picard et al., 2015). The dried weight of each oven-dried sample was
recorded after removal from the oven. The total dried weights (TDWc) of each component of
the sample trees were calculated based on the total fresh weights of each component (TFWc)
measured in the field and the ratios of dried weight to fresh weight (SDWc/SFWc) calculated
for each component in the laboratory (Huynh et al., 2021). The T DWc he TDWc was
calculated using the e formula of (Equation 1).

SDWc
TDwc=TFWc ∗100 0 Equation 1
SFWc

Where: TDWc is the total dry weight of a component c (stem, branches and leaves); TFWc is
the total fresh weight of this component measured in the field; SDWc and SFWc are the dry
sample weight and fresh weight samples for this component. The total aboveground biomass
of each sampled tree was calculated as the sum of total dry weight.
TDW tree =TDWS+TDWB+TDWL Equation 2

Where; Total Dry Weight of each tree aboveground biomass was TDWS Total dry weight of
stem, TDWB Total dry weight of brunch and TDW L Total dry weight of leave. For wood-
specific gravity estimation, for fresh stems and sample discs, diameters of the lower and upper
levels of sides were used to calculate the volume (V) using the Smalian’s formula (Sillett et
al., 2019).

π ×L
V= (D12 + D22)
8
Equation 3

Where; V represents the volume, L the length of the stem and D1 and D2 the diameters at the
lower and upper sides of the tree π = 3.14. Wood basic density (g cm-3 each disc (and tree)
were determined as the ratio of dry mass to green .

DM
Dry Mass ρ =
GV
Equation 4

Where ρ = wood basic density (g cm-3), DM = subsample oven-dried mass (g), and GV=
subsample fresh volume of each tree (cm 3) calculated from ( Equation 3) . The stem, branches
and leaves (Xiang et al., 2016) based on (equation 2).

3.4. Data Analysis

The data was arranged and organized on the Excel datasheet. The data measured both in the
field and laboratory was used to estimate the aboveground biomass components (stem, leaf,
and brunch) of the eucalyptus tree and wood density. To create biomass regression models, the
summarized data on D, H, wood density ρ , and total aboveground biomass (AGB) were
employed. One independent variable, the DBH variable, was employed in the creation of the
height prediction model. The software package Statistical Package R (version 4.2.1) was
applied to generate allometric equations and generate all of the resultant graphics.
 3.5. Allometric Equation Development

3.5.1. Model Development

The commonly used eight allometric models were developed for estimating the aboveground
biomass of E.globules tree species. According to (Uller et al., 2016) Before starting the model
development, the data were split into two sets by random sampling. The first data set
contained 75% of the data used for training or fitting model allometric models, while the
second set contained the remaining 25% of the data, which was used for validation of the
developed allometric models. Allometric models for aboveground biomass estimation were
developed using field-measured predictor variables (D, H and wood density) as independent
variables, whereas the aboveground tree biomass (AGB, kg tree-1) was used as a dependent
(response) variable.(Table 1).

Table 1:Allometric equation types tested for estimating total aboveground biomass For
E.globule s

No. Predictor Equation type Source

¿
M1 D Y =a+bD 2 (Brown et al., 1989)
2
M2 D Y =a+bD+ c D
M3 D ln ( Y ) =a+b ln ⁡(D) (Brown, 1997)
M4 D,H ln ( Y ) =a+bln (D)+cln¿) (Djomo et al., 2010)
M5 D, ρ ln ( Y ) =a+bln ( D )+ c ln ⁡( ρ)⁡
M6 D, H ln ( Y ) =a+bln ( D )+ cln(H )+ d ln ⁡( ρ)⁡ (Nelson et al., 1999)
M7 D,H, ρ ln ( Y ) =a+b ln ¿) (Chave et al., 2005)
M8 D, H, ρ ln(Y)= a+ bln( D 2∗H∗ρ)

Where Y: is the predicted tree biomass value in kg and a, b, c and d are the fitted parameters,
ln: natural logarithm H: is total height D: is diameter at the breast height ρ : is Woody
Density
3.5.2. Model Selection and Validation

All the models listed above were tested based on different parameters. Models with a
prediction power of total aboveground biomass using the number of variables (diameter,
height, and wood density) were selected using three parameters. First, we develop equations
using only diameter as a predictor. Then we stepwise include height and density and analyze
their effects on the predictive quality of the models. The best-fit allometric model was selected
based on goodness-of-fit statistics.

The model with the lowest Akaike Information Criterion (AIC), residual standard error (RSE),
and highest c The adjusted R2 coefficient of determination (AdjiR2) values was selected as the
best-fit model (Chave et al., 2005; Kuyah, 2012; Mugasha et al., 2013). Diagnostic plots of
observed values versus fit and residual trends were also used to assess model performance. For
each of the models tested, the following indicators of goodness of fit are reported:

1. Adjusted R2: corrects the coefficient of determination by accounting for an increasing

number of independent variables.

RSS
Adj .R2 n− p Equation 5
¿ 1−
TSS ( n−1 )

Where RSS =Residual sun squares TSS= total sum of squares n = number of samples p=
number of observation

2. Residual standard error of estimate (RSE): square root of the residual variance around
the regression function.

RSE = √∑ ( y−^y ) /df

2
Equation 6
y: The observed value; ŷ: The predicted value; df: The degrees of freedom, calculated
as the total number of observations – total number of model parameters.
 3. Akaike Information Criterion (AIC): the measure of goodness of fit of a regression
model. The regression equation with the lowest AIC value is the best estimator
AIC = −2 ln ( L ) +¿ 2 p ¿Equation 7

Where p is the number of parameters in the regression model, .L the likelihood of the data
under the according regression model ln: natural logarithm

In forest biomass studies, the error variances for the allometric nonlinear equations based on
arithmetical units of measurement are not constant over all observations (heteroscedasticity) in
most cases. Using log-transformed data for linear regressions when estimating the parameters
in nonlinear models is one of the most commonly used methods for eliminating the influence
of heteroscedasticity. Due to this, the allometric equations were performed using a natural
logarithmic transformation because the transformation corrects the data for heteroscedasticity
(Astivia & Zumbo, 2019).

The predicted values are on the x-axis. If the scatter plot does not show any pattern, it means
the data are free from heteroscedasticity, and vice versa. With this transformation, the
variances of the entire data were corresponding to meet the prerequisites of the linear
regression. However, log transformation introduces a systematic bias on the original scale that
is generally corrected with a correction factor CF depending on the residual standard error
(RSE).
4. Correction factor according to the predicted value using log-transformed allometric
equation is multiplied by CF to correct underestimation. The larger the RSE, the more
uncertain regression models predict biomass values, and the larger the correction factor
(Chave et al., 2005).

CF = exp ( RSE2 2 ) Equation 8

CF Correction Factor, RSE =residual standard error, exp= exponential function

Validation was carried out after the allometric equations were performed. The equations were
validated using independent data from model selection. That means the 25% set of data
randomly selected from the total sampled trees (60 trees) was used to validate data.
The predictive power of the allometric models was evaluated or validated using mean absolute
percentage error (MAPE) and Root mean square error (RMSE) (Djomo et al., 2016).

n
100 y− ^y
5. MAPE (%¿= ∑
n i=1
¿
^y
∨¿ ¿ Equation 9

n
1
6. RMSE = ∑ ( y −^y ) 2 Equation 10
n i=1

where y and yi are observed AGB of the independent dataset and ŷ and ŷi are the AGB
predicted using the allometric model being validated, n is the number of observations A
positive validation has previously been recommended when −20% ≤% Y ≤ 20% (Huang et al.,
2003).

3.6. Comparison of the Current Model with the previous General Model

For comparison with the current models, four general models were selected. These models are
frequently used in tropical forests, and in Ethiopia, the models were selected for comparison.
and all models except for the (Brown, 1997) (Table 2). Models have three predictor variables.
After selecting the allometric models, we calculated our mean AGB using the models and then
predicted the biases. When applied to all 60 of the destructive data points. Besides these, to
compare and display in further detail using the graphs of observed aboveground biomass
against the common variable D, which serves as the common variable for all of the chosen
existing general models.

Table 2:selected previously published model s

Source Predictor Models Equation

variable
(Chave et al., 2005) D,H, ⁡ρ lnY= -2.187+ 0.916 ln(D^2H ρ ) / 0.112 (D^2H ρ ) ^0.916

(Chave et al., 2014) D,H, ⁡ρ lnY= 0.0673 (D^2H ρ )^0.976

(Brown, 1997) D lnY= -2.134+ 2.53ln(D)
(Mahmood et al., D,H, ⁡ρ ln (B) = -2.46+ 2.171ln (D) + 0.367ln (H) + 0.161ln ( ρ
2019) )

lnY = logarithm of total aboveground biomass: H= total Height D=Diameter at the breast

3.7. Tree Height Prediction Model Development and Selection

The total height had been used as the dependent variable in the height-diameter relationship
model developed for smallholder plantations, and there was only one independent variable—
the DBH variable. For this study, three height prediction models that have been widely applied
in the past were assessed and developed (Table 3).

The developed and validation processes are similar in the AGB prediction model
development: model fitting (75%), and for validation (25%), data are used. The best fit model
was chosen according to the highest and lowest values of adjiR 2, RSE, and AIC, respectively,
which are statistical analyses used to assess goodness of fit used for model selctions are
sufficient when used togateher recomannded by (Chave et al., 2005).

Table 3: Allometric equation types Developed for total height prediction model for E.globules

Mod Predictor Equation type Source

No.
2
M1 DBH H=a+bDBH (Adeyemi and Ukaegbu, 2016; Koirala et
M2 DBH H=a+bDBH + c DBH
2 al., 2017)
M3 DBH ln ( H ) =a+b ln ⁡(DBH )
 Chapter 4. RESULT AND DISCUSSION

4.1. Characteristics of the Measured Trees

The data collected in the field for model development was first randomly classified into two
sections: data for model development (75%), and data for model validation (25%). The
summary of both data sets (model development and model validation) is presented in Table 4.
Accordingly, the mean values of DBH, height, wood density, and total dry aboveground
biomass in the model fitting data were 6.33 cm, 9.96 m, 0.36 g/cm3, and 6.73 kg, respectively.

Similarly, the mean values of DBH, height, wood density, and total dry aboveground biomass
for model validation data were 6.24 cm, 9.71 m, 0.35 g/cm3, and 6.56 kg, respectively. The
summary result showed no significant difference in data arithmetic variables between model
development and validation data. This minimized personal errors or bias in validating
allometric models developed using the fitting of development data.

Table 4:Statistical summary for tree parameters used in developing or fitting and validating
allometric models for aboveground biomass estimation and height prediction model

Fitting data (n=43) Validation data (n=17)

Tree parameters Mean. Max. Min. Sd. Mean. Max. Min. Sd.

DBH (cm) 6.33 12 2 2.97 6.24 12 2.1 3.21

Height (m) 9.96 14.2 5.2 2.37 9.71 14.2 5 2.97

Wood density (gcm-

3
) 0.36 0.71 0.18 0.12 0.35 0.64 0.09 0.13

Dry biomass (kg) 6.73 23.46 0.67 6.41 6.56 23.94 0.68 6.67

The correlation between measured tree parameters (height, diameter, and wood density) and
total aboveground biomass is presented (Figure 2). As shown in the graph, the total
aboveground biomass had a positive correlation with total tree height and diameter at the
breast. However, the wood density didn’t show either an increasing or decreasing trend against
the aboveground biomass (Stegen et al., 2009).
Figure 2:Scatter plot of tree
aboveground tree biomass (AGB)
vs. potential predictor
variables (diameter,
height, and wood-specific
density).

4.2. Model development and selection

In this study, eight allometric equations were assessed and developed to estimate the total
aboveground biomass of E.globules tree plantations (Table 5). Two of the models (M1 and
M2) of the eight allometric equations were created using normal data, while the remaining six
model equations (M3–M8) were created using log-transformed data.

The results showed that coefficient determination (adji R 2) values varied from 0.866 to 0.94,
which implies the aboveground biomass could be explained by 94% with the developed
model. The residual standard error ranges from 0.230 to 4.98; the Akakike Information
Criteria ranges from 1.297 to 264.04; and the correction factor ranges from 1.027 to 1.055
were the statistically well-fitting values. Model M6 had the highest coefficient determination
(adji R2) value among the developed models, whereas model M1 had the lowest coefficient
determination (adji R2) value.

The residual standard error (RSE) value of model one (M1) was greater than that of model six
(M6). In terms of Akakike information criteria, AIC, model five (M5) had the smallest value,
model one (M1) had the greatest, and in terms of the correction factor (CF), model M6 had the
smallest value and model 4 had the largest value from the developed model (Table 5). In this
study, coefficient determination (adj R 2) values greater than or equal to 0.86%, which is near
1, indicate that all the created models fit well. It has been reported that as the adji R 2 value is
closer to 1, the model is considered the best to explain the aboveground biomass of trees
(Chicco et al., 2021) ( Table 5).

Model M2, which uses only D as a predictor variable, was found to be better with an adji R 2
value than M4 and M7, which use both D and H as predictor variables. That means M2 using
only D as a predictor could explain about 90.4% of the variability during aboveground
biomass estimation. This result agreed with the regional biomass model (Basuki et al., 2009;
Kuyah, 2012) which reported that adding height as a predictor did not improve the
aboveground biomass estimation performance of allometric models, but the adjiR 2 values
vary.

Generally, this study showed that tree diameter alone is a strong and important predictor
variable, explaining most of the variability in the AGB (86.6%–90.4%). This result is
consistent with the previous studies in Ethiopia (Mulugeta Zewdie et al., 2009 ; Birhanu
kebede &Teshome Soromessa, 2018; Amsalu Abich et al., 2019) for plantation and natural
ecosystem resepectively and also in western Kenya (Kuyah, 2012; Kuyah et al., 2013;) for
agroforestry system and pan tropical equestion ( Brown et al., 1989 ; Brown, 1997) .

All of which suggested that allometric models using D alone are strong predictor models for
aboveground biomass estimation. In addition, measuring the tree diameter is very easy, cheap,
and accurate compared to other tree parameters.

Of course, using H and D as compound predictor variables improved the overall accuracy of
allometric models. For example, unlike M2, M4, and M7, they added H as a predictor
variable, which greatly decreased the models’ RSE and AIC values. It implies that the
accuracy of models in aboveground biomass estimation improved when combining H and D as
predictor variables. Similar to what (Gadisa Demie et al.,2023) repoted for native perennial
plant the agricultural landscape of Central Ethiopia, and (Segura, 2005) reported for the
tropical humid forests in Costa Rica1, the model fit improved by adding H allometric model
development and in general model (Chave et al., 2005; Chave et al., 2014) for all types of
forest. Conclude that height is a commonly used variable, it can lead to less biased estimates
of the aboveground biomass. When model fitting includes height as a predictor variable, as
studied by ( González et al.,2013), increase the Acuracy of the model. According to
(Feldpausch et al.,2011), tree height at various sites with varying allometry is an important
independent variable, and this is also the case in the current study.

Adding woody density as the predictor variable in the D-based model, in this study, AGB was
improved and a good fit explained 91%, reducing RSE and AIC than adding height (M5). The
results agree with the previeus study on the tropical biomass equation (Chave et al., 2014 ,
2005; Djomo et al., 2010), and regional mixed species (Basuki et al., 2009; Kuyah, 2012;
Mukuralinda et al., 2021) and for mixed secondary forests (Ketterings et al., 2001; Xu et al.,
2015). Research findings demonstrated that wood density must be taken into consideration
when including site-specific equations in order to improve accuracy and yield a more accurate
estimate of biomass. When comparing DBH, height, and wood density, the influence of wood
density becomes quite noticeable.

Allometric models that included wood density as a predictor variable with diameter greatly
reduced the RSE and AIC and increased the R2 compared to total tree height. This implies
adding woody density to the model as the predictor with diameter has improved the predictive
performance of the models and aboveground estimation accuracy (M5). However, this finding
is contrary to the study of a species-specific model developed for Miombo woodlands in the
Copper Belt Province of Zambia (Kapinga et al., 2018), which pointed out that adding woody
density neither improves nor increases the predictive ability of allometric models in the DBH-
based model.
However, the findings in this study showed that woody density is a very important predictor
variable for aboveground biomass estimation.Allometric models of Eucalyptus globulus
smallholder plantation forests in the study area than the Adding height as the predictor
variable. fitting the model and increasing the accuracy of the predictive model, and similarly,
the woody density mean is in the range of 0.3–0.9 g/cm3 reported for tropical African forests
(Brown, 1997) this study the mean value 0.36 g/cm3.

Model M6 and model M8 were able to predict the variation of AGB in about 93%–94% of
smallholder plantations (Table 6). This showed that the combination of total height and woody
density with tree diameter as predictor variables highly improved the accuracy of allometric
models to estimate the AGB of E. gulobules tree planation at smallholder levels. Several
previous studied models, like the general pan-tropical equation and regional biomass equation
( Nelson et al., 1999; Chave et al., 2005,2014; Djomo et al., 2010) and in Ethiopia site and
species specific model ( Amsalu abich et al., 2021; Dessie Assefa et al., 2022). also showed
similar results that allometric models' performance in estimating aboveground biomass was
enhanced in combing three tree parameters (height, diameter and wood density).

Generally, based on the four model selection criteria, Model 6 (lnY =lnY = -2.875 + 1.243 (D)
+ 1.631 (H) + 0.665 (ρ), which combined three predictor variables, was found to be the first-
ranked best allometric model. Model six recorded the highest adji R 2 (0.94) and the lowest
RSE (0.230) and CF (1.027) values of all tested models. Similarly, Model M8 (lnY = -
1.602+0.831ln(D^2 H ρ)) was the second-ranked best model next to Model M6 and had the
second highest R2 (0.931) and the lowest RSE (0.247), AIC (5.58) and CF (1.031) next to
Model M6.

The finding of this study that the allometric model, which uses three predictor variables (D, H,
and ρ), gave the best performance (M6) agreed with the findings of the of the site-specific
model for peat swamp forests in Indonesia and Combretum-Terminalia woodlands in Ethiopia
(Nugroho, 2014; Amsalu Abich et al., 2021) Respectively, and for pan-tropical biomass
equations ( Nelson et al., 1999; Chave et al., 2005).

Showed The majority of the variation in the increase in aboveground tree biomass in the
current study can be explained by DBH when combined with wood density and total height.
Thus, models using DBH along with additional predictors like H and ρ were considered to
have the highest fit and predictive power in this study since they produced lower AIC, RSE
CF,and higher R2, explaining a significant amount of the variability in the aboveground
biomass, as well as larger adji R2 values and significant ρ and H parameter estimates.

Table 5:Developed allometric model to estimate total aboveground biomass

Mod. Equation Type A B C D Adj.R2 RSE AIC CF

2
M1 Y= a+ bD -12.782 4.28 0.866 4.98 264.04 -
2
M2 Y=aD+ bD 14.322 82.361 17.5 0.904 4.214 250.63 -
M3 ln(Y)=a+bln(D) -0.965 1.847 0.878 0.328 30.11 1.055
M4 ln(Y)=a+bln(D)+cln(H) -2.854 1.201 1.327 0.897 0.302 23.88 1.047
M5 ln(Y)=a+bln(D)+cln( ρ ) -0.608 2.013 0.577 0.910 0.282 18.15 1.041
M6 ln(Y)=a+bln(D)+cln(H)+dln(" ρ -2.875 1.243 1.631 0.665 0.940 0.230 1.297 1.027
)
M7 ln(Y)=a+ bln(D^2 *H) -2.047 0.747 0.895 0.304 23.60 1.047
M8 ln(Y)=a+bln(D^2 *H* " ρ ) -1.602 0.831 0.931 0.247 5.58 1.031

Where: Total aboveground biomass (Y, kg); Diameter at the breast height (D, cm); Total tree
height (H,m); wood density ( ρ , gcm-3); a, b, c and d are the model’s fitted parameters;
Coefficient of determination (Adj R2); Residual standard error (RSE); Akaike Information
Criterion (AIC), Correction factor (CF).

4.3 Model Validation

To rank the final best model out of the selected models, MAPE and RMSE were calculated
using a 25% independent data set of trees (Table 6). The validation result showed that the
MAPE% value of selected models ranged from 15.71% to 80.64%. Accordingly, M6,
combining diameter, height, and wood density as (lnY=a+ bln D+clnH+dlnρ) showed the
lowest MAPE (15.71%).

The expected value of MAPE is zero and a percentage different from zero shows the deviation
of predicted mean AGB by the given model relative to the observed mean AGB . Hence, M6 is
the best model of all the tested models with the lowest prediction error relative to the observed
AGB. Similarly, Since the relative value of the residual is squared, RMSE offers a strong tool
for both validating equations and determining the final ranking in allometric models reported
by (Dijomo et al., 2016). The validation results still showed that M6 was found to be the best
model with the lowest RMSE (2.88).

Table 6: Mean absolute percentage error (MAPE(%) and relative root mean square error
(RSME) for AGB estimation allometric equations developed using a 25% (17) independent
set of data.

Mod Equations MAPE (%) RMSE (kg)

.
M1 Y=-12.782 + 4.28D^2 80.64 5.44
M2 Y=14.322+82.361D+ 17.5D^2 26.48 3.57
M3 ln(Y)=-0.965+1.847ln(D) 25.32 4.31
M4 ln(Y)= -2.854+1.201ln(D)+ 1.327ln(H) 25.71 4.30
M5 ln(Y)=-0.608+2.013ln(D)+ 0.577ln( ρ ) 18.52 2.94
M6 ln(Y)=-2.875+1.243ln(D)+ 1.631ln(H)+ 0.665ln( ρ
) 15.71 2.88
M7 ln(Y)=-2.047+0.747ln(D^2 H) 25.40 4.21
M8 ln(Y)=-1.602+0.831ln(D^2 H " ρ ) 19.20 3.14

Note: Total dry aboveground biomass (Y), Diameter at breast height (D), total tree height (H)

The diagnostic scattered residual plots against the predicted AGB for the three selected models
are presented in Figure 5. The diagnostic scattered residual plots against the predicted AGB of
the three selected best-fitted AGB models for our study site-specific model indicate an even
spread of residuals above and below the zero line with no systematic trend (Figure 3).

This suggests that natural log-transformed multiple linear regressions are effective in
stabilizing errorvariance,and this shows model appropriateness for reducing heteroscedasticity.
The best-selected model (M6) errors in the residual plots were uniformly distributed between
±6, which implies that the mean difference between observed and predicted could be zero
(Packard, 2013)(Figure 3). This indicates that the selected models (M6) had a good fit .

Figure 3: Scatter plots of residual values versus predicted value of the three best allometric fit
models Predicted value in log scale

4.4. Comparison of the Current Model with the Existing General Models

The current site-specific allometric aboveground biomass models developed in this study (M6)
were compared with the four existing common general allometric models (Table 7). The result
showed that the predicted AGB, with the current general-compared models, ranged from 10.56
kg to 18.13 kg. Similarly, the relative error of the current study model (M6) and general
developed models ranged from -2.11% to 27.5%. Hence, the current study model (M6)
recorded the lowest relative bias error (-2.11%) of the other general models.

This shows that using general models as AGB estimation models for eucalyptus plantations in
the study area could either highly underestimate or overestimate the AGB. This finding
implies that the use of the general model is associated with biased estimates of aboveground
biomass (AGB) when they are applied at a particular site and species. (Henry et al., 2011;
Mugusha et al., 2013).

Hence, site-specific allometric equations are very important to boost the prediction accuracy
of models. According to (Huang et al., 2003). A bias of less than ± 10% a is considered
acceptable, which is recommended as the bias measure of systematic devotion of model
prediction from measured data. The average deviation for individual trees of the previously
published models is always higher than that of model M6 (Table 8). As presented in Table 8,
the allometric equations of (Chave et al., 2005; 2014), and (Mahmood et al., 2019) showed an
underestimation of AGB compared to the observed AGB, whereas (Brown,1997)
overestimated the predicted AGB value.

Table 7: comparison of the current best selected model with the existing models using the
whole measured data (N=60).

Models sources Mean biomass(kg/Tree)

Measured Predicted Bias (%)
Current selected model 14.22 13.92 -2.11
(Chave et al., 2005) 12.53 -11.88
Chave et al., 2014) 10.56 -25.74
(Brown, 1997) 18.13 27.50
Mahmood et al, 2019) 12.26 -13.78

The TAGB versus DBH plot graph (Figure 4) showed that the current model (M6)
approximately estimates by -2.11% of the observed data. The purple cross represented the -
11.88% underestimation of AGB in the (Chave et al., 2005) model: lnY= -2.187+ 0.916
ln(D^2H ρ ) / 0.112 (D^2H ρ ) ^0.916. A green cross was underestimated by -25.74% using the
model (Chave et al., 2014), lnY= 0.0673(D^2H ρ)^0.976. In addition to overestimating AGB by
(Brown, 1997) 27.50%, the model, lnY= lnY=−2.134+ 2.53ln(D).According to (Mahmood et
al., 2019) model, the red curve represents a -13.78% underestimate of AGB ln (B) = -2.46+
2.171ln (D) + 0.367ln (H) + 0.161ln (ρ).

According to this findings (Chave et al., 2005) model seems to perform better when compared
to the other three general models and becames underestimation. The Underestimation and
overestimation were found by the four general models because models should be applied to
specific DBH sites and species ranges (Chave et al., 2005; 2014; Henry et al., 2011; Mugusha
et al., 2013).The woody density and height also vary from site to site and species to species
(Dutc et al., 2018).

In general, this finding concludes that there were variations in the aboveground biomass using
the general equation, the site-specific equation, and the field data calculation. When using
general allometric models, the equations' estimations varied and the errors increased (Figure
4).Similar to this, a general allometric model is created for a range of sites that could expand
over a wider geographic area, but it excludes factors like soil type, altitude, and climate that
could affect the biomass of trees. However, as the name suggests, the site-specific approach is
highly specialized for a specific site in a certain environmental condition in particular, such as
stand growth, age, and management.As a result, using the current models is appropriate for
estimating the aboveground biomass of smallholder plantations of E. globuluse. In In addition,
when there is no available height and woody density data using model M2, which uses only
diameter as a predictor variable, it is more preferable to estimate the AGB of eucalyptus
plantations in the study area.
Figure 4: Comparison of the current model with previously general equation used to estimate
aboveground biomass of smallholder plantation of Eucalyptus globulus tree.

4.5. Height Prediction Model Development and Selection

The correlation between the diameter and the height of the scatter plot showed that the trend
increased with the diameter (see figure 5). The coefficient determination (adjiR 2) values for
models ranged from 0.81 to 0.86 (Table 8). This shows that the developed models were well
fitted, and the dependent variable height of smallholder plantations was predicted by 81% to
86% with the independent variable. This result also agrees with (Liu et al., 2017), pointed out
that the diameter at breast height used as the independent variable in the allometry equation of
tree height can more accurately represent the change in tree height.
Figure 5: scatter plots of Total height versus the potential of predictor Diameter

The RSE ranged from 0.102 to 1.09, and the AIC ranged from -78. to 142.82. The largest RSE
was recorded by model 1, while the lowest model, M3, based on the three good fitnesss
(Table 9), ranked the best model, M3, log models, which is rankly the first selected model
from the three developed models. Followed by model M2, whereas model M1 was the last-
ranked model.

The selected height-diameter models are presented in Table 8. According to the criteria used
for selecting the most suitable prediction model (M3), the logarithmic height-diameter
functions performed better than the two models for tree height predictions in the smallholder
plantation. Although the adjiR2 value (0.84) was lower than the values reported by (Adeyemi
and Ukaegbu, 2016) between 0.75 and 0.9 for Gmelina arborea plantations, (Caetano et al.,
2014), ranging from 0.82 to 0.95 for brazil nut tree and in Ethiopian Geda forest ranging from
0.88 – 0.89 for Cupressus lusitanica (Mengsha Tsega et al., 2018). This finding suggests that
the height was influenced by the type of forest, the site, and the age of the stand. Each type of
species requires a different height diemter relation model.

The best-fit selected model had smaller RSE and AIC values. This result implies greater
accuracy for tree height prediction in the study area. According to (Vieilledent et al., 2012), a
basic height-diameter allometry is required to accurately estimate biomass and carbon stocks
from plot inventories, when biomass allometric models for a specific forest site are not
present. This study supported these findings and developed the logarthem height model for
small plantations found in the study area and similar geographical areas to estimate accurate
aboveground biomass.

Table 8: Developed Allometric model to prediction total height of E.globuls smallholder

plantation

Mod. Equation Type A B C d Adj.R2 RSE AIC

M1 H=a+bD^2 5.29 0.74 0.81 1.02 142.82
M2 H=a+D+D2 10.04 14.73 -3.71 0.86 0.88 128.85
M3 ln(H)=a+bln(D) 1.40 0.5 0.84 0.10 -78.6
lnH; natural logarithm of height H; Height D; Diameter M model a, b, c and d are the model’s
fitted parameters; RSE is the residual standard error; Adj R 2 is the coefficient of determination,
AIC is the Akaike Information Criterion, CF is Correction factor and n = 43.

For validation, we used the scattered plot residual and observed plots against predicted height.
for the best-selected models is presented in Figure 6. The figure shows that the mean error
(observed minus predicted) was approximately zero, which implies that the model was a good
fit. Also, the observed versus predicted scatter plot shows that the points are close to each
other, which indicates the model was a good fit and is equally distributed (Koirala et al.,
2017).

Figure 6: residual and observed plots against predicted height for the best-fit model for this
study
 Chapter 5. CONCLUSION

This study developed eight site-specific aboveground biomass estimation models by using 60
destructive sample trees for E.globuluse plantation. The finding showed that using tree
diameter as a single predictor variable highly decreased the aboveground prediction efficiency
of allometric models. Similarly, combining tree diameter and tree height still resulted in lower
accuracy in estimating the AGB of the plantation. However, combining the three predictor
variables (diameter, height, and wood-specific density) to develop allometric models highly
improved the aboveground biomass estimation of the models. According to the developed
measuring criteria, the best-fitted model is the one that combines the three predictor variables
in the form of ln(Y)=-2.875+1.243ln(D)+1.631ln(H)+0.665ln( ρ ).To estimate the aboveground
biomass of the Eucalyptus plantation in Wogera district. This developed model still showed
better aboveground biomass estimation potential compared with the other four existing general
models, which have been used by many researchers in different parts of the world. On the
other hand, the best-developed height prediction model was: ln(H) = 1.400 + 0.5 ln(D), which
could help to estimate tree height using measured tree diameter. Because, in reality, measuring
tree height, especially in dense plantation forests, is very challenging and time-consuming.
Hence, such diameter-height relation allometric models are very important to estimate the
height of each tree with its measured diameter at breast height.
 Chapter 6. RECOMMENDATION

Based on this study's findings, researchers are recommended to estimate the aboveground
biomass of such plantations. This can avoid the underestimation or overestimation of biomass
production, which could affect both the grower and buyers during the selling and buying of
plantations. In developing allometric models for biomass estimation, it is more recommended
to use the three variables (tree diameter, height, and wood density). Of course, it is not always
easy to measure and use all the required tree variables. Hence, selecting and using better
allometric model equations that use diameter only as a predictor variable is recommended.
 REFFERENCE

Abebe Birara Dessie, Mirie Abate Tadie, Taye Melese Mekie. (2019). Eucalyptus : The
Popular Exotic Tree Crop in Ethiopia. Acta Scientific Agriculture. 3:50–56.
https://doi.org/10.31080/ASAG.2019.03.0607.

Adeyemi, A.A. and Ukaegbu, O. C. (2016). Tree height-diameter and yield functions for
Gmelina arborea (roxb.) stand in edondon gmelina plantation, cross river state, nigeria.
Journal of Research in Forestry. 8(2): 126–144.

Admassu, A., Teshome, S., & Tura, B. (2022). Allometric Equation for Aboveground Biomass
Estimation for Selected Trees Shrubs in Gesha-Sayilem Moist Afromontane Forest.
Journal of Forest Research, 11(4): 1–8. https://doi.org/10.35248/2168-9776.22.11.323.

Alemayehu Asabeneh &, & Melka Yoseph. (2022). Small scale eucalyptus cultivation and its
socioeconomic impacts in Ethiopia : A review of practices and conditions. Trees, Forests
and People. (8), 100269. https://doi.org/10.1016/j.tfp.2022.100269.

Amsalu Abich, Tadesse Mucheye, Mequanent Tebikew, Yohannis Gebremariam, Asmamaw

Alemu. (2019). Species-specific allometric equations for improving aboveground
biomass estimates of dry deciduous woodland ecosystems. Journal of Forestry Research,
30(5), 1619–1632.

Amsalu Abich, Asmamaw Alemu, Yohannis Gebremariam, Tadesse Mucheye, Kassaye

Gurebiyaw, Meseret Kassie. (2021). Allometric models for predicting aboveground
biomass of Combretum-Terminalia woodlands in Amhara, Northwest Ethiopia. Trees,
Forests and People. 5: 100122.

Antonio, N., Tomé, M., Tomé, J., Soares, P., & Fontes, L. (2007). Effect of tree, stand, and
site variables on the allometry of Eucalyptus globulus tree biomass. Canadian Journal of
Forest Research, 37(5), 895–906.

Astivia, O. L. O., & Zumbo, B. D. (2019). Heteroskedasticity in multiple regression analysis:

What it is, how to detect it and how to solve it with applications in R and SPSS. Practical
Assessment, Research and Evaluation, 24(1), 1–16.
Author, V., Vaudry, R., Andriamanohisoa, S. F. D., Rakotonarivo, O. S., Razafindrabe, H. N.,
Rakotoarivony, C. B., Ebeling, J., Vieilledent, G., Vaudry, R., Andriamanohisoa, S. F.
D., Rakotonarivo, O. S., & Randrianasolo, H. Z. (2012). A universal approach to estimate
biomass and carbon stock in tropical forests using generic allometric models. The
Ecological Society of America A, 572=583.

Banerjee, K., Sahoo, C. K., Bal, G., & Paul, R. (2022). Allometric model based on
dendrometrical measurements in mangroves of western Bay of Bengal. Tropical Ecology,
63(3), 351–364. https://doi.org/10.1007/s42965-021-00215-9

Barbati, A., Brusselen, J. Van, & Lindner, M. (2007). The Role of Forests in Carbon Cycles,
Sequestration, and Storage: "Climate Change Mitigation, Forest Management and Effects
on Biological Diversity.News letter.5.

Basuki, T. M., Leksono, B., Baral, H., Andini, S., Wahyuni, N. S., Artati, Y., Choi, E., Shin,
S., Kim, R., Yang, A., Samsudin, Y. B., & Windyarini, E. (2022). Allometric Equations
for the Biomass Estimation of Calophyllum inophyllum L . in Java , Indonesia. 1–15.

Basuki, T. M., van Laake, P. E., Skidmore, A. K., & Hussin, Y. A. (2009). Allometric
equations for estimating the above-ground biomass in tropical lowland Dipterocarp
forests. Forest Ecology and Management, 257(8), 1684–1694.
https://doi.org/10.1016/j.foreco.2009.01.027

Belayneh Dessie, Mintesinot Taye1, Zablon Adane, Ayana Jember (2022). Analysis of soil
carbon and income over Acacia decurrens and Eucalyptus globulus land uses in the
highlands of Ethiopia. Journal of Environmental Studies and Sciences, 1-12.
https://doi.org/10.1007/s13412-022-00784-8.

Birhanu kebede & Teshome Soromessa. (2018). Allometric equations for aboveground
biomass estimation of Olea europaea L. subsp. cuspidata in Mana Angetu Forest.
Ecosystem Health and Sustainability, 4(1): 1–12
https://doi.org/10.1080/20964129.2018.1433951.

Biruk Ketsela Hailemicael. (2012). The Contribution of Eucalyptus Woodlots to the

 Livelihoods of Small Scale Farmers in Tropical and Subtropical Countries with Special
Reference to the Ethiopian Highlands.Master Thesis. In Swedish University of
Agricultural Sciences Faculty of Forest Sciences Department (Issue 89), 1654-1367.

Blake, S. T., Nusa, E., Sadono, R., Wardhana, W., Yudha, P., Putra, A., & Idris, F. (2021).
Allometric Equations for Estimating Aboveground Biomass of. jtfm.27:24–31.
https://doi.org/10.7226/jtfm.27.1.24

Bosques, D. E. N., & Secundarios, T. (2019). species-specific biomass equations for small-
size tree species in secondary tropical forests. Tropical and Subtropical Agrosystems, 22:
735–754.

Brown. (1997). Estimating Biomass and Biomass Change of Tropical Forests : Food and
Agriculture Organization of the United Nations, Rome .

Brown S, Gillespie AJR, Lugo AE (1989) Biomass estimation methods for tropical forests
with applications to forest inventory data. For Sci 35(4):881–902.

Caetano, A. C., Ribeiro, S. C., de Macedo, R. L. G., & Dias, B. A. S. (2014). Height-diameter
equations for Brazil nut intercropped with rubber tree in the South of Minas Gerais.
Floresta, 44(3), 497–504. https://doi.org/10.5380/rf.v44i3.32514.

Calfapietra, C., A. Barbati, L. Perugini, B. Ferrari, G. Guidolotti, A. Quatrini, and P. Corona.

2015. Carbon stocks and potential carbon sequestration of different forest ecosystems
under climate change and various management regimes in Italy. Ecosystem Health and
Sustainability 1(8):25. http://dx.doi.org/10.1890/EHS15-0023.

Canadell, J. G., & Raupach, M. R. (2008). Managing forests for climate change mitigation.
Science, 320 (5882), 1456–1457. https://doi.org/10.1126/science.1155458

Chave J, Andalo C ., Brown S, Cairns MA, Chambers JQ., Eamus D, Fo¨lster H., Fromard F.,
Higuchi N, Kira T., Lescure JP., Nelson BW, Ogawa H, Puig H, Rie´ra B, Yamakura T
(2005) Tree allometry and improved estimation of carbon stocks and balance in tropical
forests. Oecologia 145:87–99
Chave J, Re´jou-Me´chain M, Bu´rquez A, Chidumayo E, Colgan MS, Delitti WB, Duque A,
Eid T, Fearnside PM, Goodman RC, Henry M, Martı´nez-Yrı´zar A, Mugasha WA,
Muller-Landau HC, Mencuccini M, Nelson BW, Ngomanda A, Nogueira EM, Ortiz-
Malavassi E, Pe´lissier R, Ploton P, Ryan CM, Saldarriaga JG, Vieilledent G (2014)
Improved allometric models to estimate the aboveground biomass of tropical trees. Glob
Change Biol 20:3177–3190.

Chicco, D., Warrens, M. J., & Jurman, G. (2021). The coefficient of determination R-squared
is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis
evaluation. PeerJ Computer Science, 7: 1–24. https://doi.org/10.7717/PEERJ-CS.623

Condit, R. (2008). Methods for estimating above ground biomass of forest and replacement
vegetation in the tropics. Center for Tropical Forest Science Research Manual.73pp.

Damena Edae Daba and Teshome Soromessa. (2019). The accuracy of species-specific
allometric equations for estimating aboveground biomass in tropical moist montane
forests: case study of Albizia grandibracteata and Trichilia dregeana. Ecosystem Health
and Sustainability. 5(1), 86–97. https://doi.org/10.1080/20964129.2019.1591169

Dagm Fiker Abate . (2020). An ecological-economic approach to assess impacts of Eucalyptus

Woodlot Expansion in Agroforest Landscapes of Northern Ethiopia, Master Degree, de
Agronomia, Universidade de Lisboa Brigite, Protugal 49pp.

DeLucia, E. H., Maherali, H., and Carey, E. V. (2000). Climate-driven changes in biomass
allocation inpines. Global Change Biology, 6(5):587–593. https://doi.org/10.1046/j.1365-
2486.2000.00338.x.

Dessie Assefa., Asnake Mekuriaw , Mathias Tesfaye, Habtamu Sewnet,Beyene Belay. (2022).
Mapping of Prosopis juliflora rate of expansion and developing species - specific
allometric equations to estimate its aboveground biomass in the dry land of Ethiopia.
Modeling Earth Systems and Environment, 0123456789. https://doi.org/10.1007/s40808-
022-01495-3

Djomo, A. N., Ibrahima, A., Saborowski, J., & Gravenhorst, G. (2010). Allometric equations
 for biomass estimations in Cameroon and pan moist tropical equations including biomass
data from Africa. Forest Ecology and Management, 260(10): 1873–1885.
https://doi.org/10.1016/j.foreco.2010.08.034

Djomo, A. N., Picard, N., Fayolle, A., Henry, M., Ngomanda, A., Ploton, P., McLellan, J.,
Saborowski, J., Adamou, I., & Lejeune, P.(2016).Tree allometry for estimation of
carbonstocksinAfricantropical forestsForestry.89(4):446–455.
https://doi.org/10.1093/forestry/cpw025.

Dutcă , Ioan , Mather , Richard , Blujdea , Viorel N . B ., Ioraș , Florin , Olari , Mănăilă and
Abrudan , Ioan Vasile ( 2018 ) Site-effects on biomass allometric models for early
growth plant. 1–27.

Ebuy, J., Lokombe, J. P., Ponette, Q., Sonwa, D., & Picard, N. (2011). Allometric equation for
predicting aboveground biomass of three tree species. Journal of Tropical Forest Science,
125–132.

Ekoungoulou, R., Nzala, D., Liu, X., & Niu, S. (2018). Tree Biomass Estimation in Central
African Forests Using Allometric Models. 209–237.
https://doi.org/10.4236/oje.2018.83014

Fauzi, M. I. D., & Rohadi, D. (2015). Complex regulatory frameworks governing private
smallholder tree plantations in Gunungkidul District , Indonesia. Forest Policy and
Economics Journal. 59(1–6). Www.Elsevier.Com/Locate/Forpol Complex.

Fayolle, A., Doucet, J., Gillet, J., Bourland, N., & Lejeune, P. (2013). Forest Ecology and
Management Tree allometry in Central Africa : Testing the validity of pantropical multi-
species allometric equations for estimating biomass and carbon stocks. forest ecology and
management, 305:29–37. https://doi.org/10.1016/j.foreco.2013.05.036

Feldpausch T. R., L. Banin1,O. L. Phillips1, T. R. Baker1, S. L. Lewis1, C. A. Quesada,K.

Affum-Baffoe, E. J. M. M. Arets, N. J. Berry1, M. Bird, E. S. Brondizio, P. de Camargo,
J. Chave, G. Djagbletey, T. F. Domingues,M. Drescher, P. M. Fearnside,M. B. Franc¸, N.
M. Fyllas, G. Lopez-Gonzalez, A. Hladik, N. Higuchi, M. O. Hunte, Y. Iida, K. A. Salim,
 A. R. Kassim, M. (2011). Biogeosciences, 8: 1081–1106.

Gadisa Demie, Mesele Negash, Zerihun Asrat, Lojka Bohdan.(2023). Species specific and
multi-species allometric models for estimating aboveground biomass of native perennial
plant species grown in the agricultural landscape of Central Ethiopia. Agroforest Syst 98,
179–196 (2024). https://doi.org/10.1007/s10457-023-00898-0

Getahun Amare. (2003). Eucalyptus farming in Ethiopia: the case for eucalyptus woodlots in
the Amhara region.

Gibbs, H. K., Brown, S., Niles, J. O., & Foley, J. A. (2007). Monitoring and estimating
tropical forest carbon stocks: making REDD a reality. Environmental Research Letters,
2(2007): 045023. https://doi.org/10.1088/1748-9326/2/4/045023

Gonzalez-Benecke, C. A., Fernández, M. P., Albaugh, T. J., Ahumada, R., Bown, H. E.,
Gayoso, J., Gerding, V., Mardones, O. B., Rodríguez, A. R., & Rubilar, R. (2021).
General above-stump volume and biomass functions for Pinus radiata, Eucalyptus
globulus and Eucalyptus nitens. Biomass and Bioenergy, 155: 106280.

Gonzalez-garcía, M., Hevia, A., Majada, J., Barrio-anta, M., & Gonza, M. (2013). Above-
ground biomass estimation at tree and stand level for short rotation plantations of
Eucalyptus nitens ( Deane & Maiden ) Maiden in Northwest. Biomass and Bioenergy.54:
147-157. https://doi.org/10.1016/j.biombioe.2013.03.019.

Haileab Zegeye. (2010). Environmental and Socio-Economic Implications of Eucalyptus in

Ethiopia. Eucalyptus Species Management, History, Status and Trends in
Ethiopia.Proceedings from the Congress Held in Addis Ababa.September 15th-17th,
2010, September 2010, 184–205.

Hartono, R., & Kusmana, C. (2020). Allometry of Biomass and Carbon Stock of Planted
Eucalyptus grandis Forest in Toba Highland. International Conference of Science,
Technology, Engineering, Environmental and Ramification Researches, 1, 129–131.
https://doi.org/10.5220/0010102401290131

Henry, M., Picard, N., Trotta, C., Manlay, R., Valentini, R., Bernoux, M., & Saint André, L.
 (2011). Estimating tree biomass of sub-Saharan African forests: a review of available
allometric equations . Silva Fennica 45(3B): 477–569.

Herbohn John , Medrilzam Medrilzam. (2014). A Small-Scale Forestry Perspective on

Constraints to Including REDD in International Carbon Markets. Small-scale Forestry.
https://doi.org/10.1007/s11842-002-0001-3

Houghton, R. A. (2005). Tropical deforestation as a source of greenhouse gas emissions. ,

Tropical deforestation and climate change / edited by Paulo Moutinho and Stephan
Schwartzman. -- Belém - Pará - Brazil : IPAM - Instituto de Pesquisa Ambiental da
Amazônia ; Washington DC - USA : ISBN: 8587827-12-X.

Huang, S., Yang, Y., Wang, Y., Amaro, A., Reed, D., & Soares, P. (2003). A critical look at
procedures for validating growth and yield models. Modeling Forest Ecosystem. 0271:
(271–292).

Huynh, T, Lee, D. J., Applegate, G., & Lewis, T. (2021). Field methods for above and
belowground biomass estimation in plantation forests. MethodsX. 8:101192.
https://doi.org/10.1016/j.mex.2020.101192

Jagger Pamela, Pender John (2003).The role of trees for sustainable management of less-
favored lands: the case of eucalyptus in Ethiopia .Forest Policy and Economics 5; 83 -95.

Jibrin, A., & Abdulkadir, A. (2015). Allometric Models for Biomass Estimation in Savanna
Woodland Area, Niger State, Nigeria. International Journal of Environmental. 9(4): 275–
283.

Kanninen, S. &. (2005). Allometric models for tree volume and total aboveground biomass in
a tropical humid forest in Costa Rica. Biotropica, 37(1):2–8.
https://doi.org/10.1111/j.1744-7429.2005.02027.x

Kaonga Martin &, & Smith Bayliss. (2010). Allometric models for estimation of aboveground
carbon stocks in improved fallows in eastern Zambia. Agroforestry Systems, 78(3), 217–
232. https://doi.org/10.1007/s10457-009-9253-7
Kapinga, K., Syampungani, S., Kasubika, R., Yambayamba, A. M., & Shamaoma, H. (2018).
Species-specific allometric models for estimation of the above-ground carbon stock in
miombo woodlands of Copperbelt Province of Zambia. Forest Ecology and Management,
417(February), 184–196. https://doi.org/10.1016/j.foreco.2018.02.044

Kassegn Asnakew Seyoum, Bekalu Melis Alehegn, Asmamaw Alemu Abtew. (2021). Effect
of Eucalyptus Globulus Woodlot Plantation on Selected Soil Physico-Chemical
Properties , Wheat Yield in Wogera District , Amhara Region , Ethiopia. 6(2), 39–49.
https://doi.org/10.11648/j.ajset.20210602.14

Kebede Gizachew. (2017). Expansion of Eucalypt Woodlot and Its Factors in. 66, 163–180.

Kenzo, T., Ichie, T., Hattori, D., Itioka, T., Handa, C., Ohkubo, T., Kendawang, J. J.,
Nakamura, M., Sakaguchi, M., Okamoto, M., Tanaka-oda, A., Sakurai, K., Ninomiya,
I., .(2009). Development of Allometric Relationships for Accurate Estimation of above-
and Below- Ground Biomass in Tropical Secondary Forests in Sarawak , Malaysia.
Journal of Tropical Ecology. 25(4): pp. 371-386 (Jul.,
https://doi.org/10.1017/S0266467409006129

Ketterings, Q. M., Coe, R., Noordwijk, M. Van, Ambagau, Y., & Palm, C. A. (2001).
Reducing uncertainty in the use of allometric biomass equations for predicting above-
ground tree biomass in mixed secondary forests. Forest Ecology and Management. 146,
199–209.

Koirala, A., Kizha, A. R., & Baral, S. (2017). Modeling Height-Diameter Relationship and
Volume of Teak (Tectona grandis L. F.) in Central Lowlands of Nepal. Journal of
Tropical Forestry and Environment, 7(1), 28–42. https://doi.org/10.31357/jtfe.v7i1.3020

Kuyah, S. (2012). Allometric Equations for Estimating Tree Biomass in Agricultural

Landscapes in Western Kenya. Phd Thesis, Jomo Kenyatta University of Agriculture
Kenya .

Kuyah, S., Dietz, J., Muthuri, C., Noordwijk, M. Van, & Neufeldt, H. (2013). Allometry and
partitioning of above- and below-ground biomass in farmed eucalyptus species dominant
 in Western Kenyan agricultural landscapes. Biomass and Bioenergy, 55, 276–284.
https://doi.org/10.1016/j.biombioe.2013.02.011

Lali Kumar t ,Priyakant Sinha, Subhashni Taylor, and Abdullah F. Alqurash.(2015). Review
of the use of remote sensing for biomass estimation to support renewable energy
generation. .15093SS.

Lin, K., Lyu, M., & Jiang, M. (2017). Improved allometric equations for estimating biomass
of the three Castanopsis carlesii H . forest types in subtropical China. New Forests, 48(1),
115–135. https://doi.org/10.1007/s11056-016-9559-z

Litton, C. M., & Kauffman, J. B. (2008). Allometric models for predicting aboveground
biomass in two widespread woody plants in Hawaii. Biotropica. 40(3): 313–320.
https://doi.org/10.1111/j.1744-7429.2007.00383.x

Löf, M., Madsen, P., Metslaid, M., & Witzell, J. (2019). Restoring forests : regeneration and
ecosystem function for the future. New Forests, 50(2), 139–151.
https://doi.org/10.1007/s11056-019-09713-0

Lorenz, K., & Lal, R. (2010). Carbon sequestration in forest ecosystems. In Carbon
Sequestration in Forest Ecosystems. https://doi.org/10.1007-978-90-481-3266-9.

Mahmood, H., Siddique, M. R. H., Costello, L., Birigazzi, L., Abdullah, S. M. R., Henry, M.,
Siddiqui, B. N., Aziz, T., Ali, S., & Al Mamun, A. (2019). Allometric models for
estimating biomass, carbon and nutrient stock in the Sal zone of Bangladesh. Forest-
Biogeosciences and Forestry, 12(1), 69.

Mandal, R. A., Yadav, B. K. V., Yadav, K. K., Dutta, I. C., & Haque, S. M. (2013).
Development of Allometric Equation for Biomass Estimation of Eucalyptus
camaldulensis: A study from Sagarnath Forest, Nepal. International Journal of
Biodiversity and Ecosystems, 1(1), 1–7.

Mekonnen Daba. (2016). The Eucalyptus Dilemma : The Pursuit for Socio-economic Benefit
versus Environmental Impacts of Eucalyptus in Ethiopia. Journal of Natural Sciences
Research, 6(19), 127–137.
Mengesha Tsega, Awoke Guadie, Zebene Lakew Teffera, Yigez Belayneh & Dongjie Niu
(2018) Development and Validation of Height-Diameter Models for Cupressus lusitanica
in Gergeda Forest, Ethiopia, Forest Science and Technology. 14:3. 138-144,
DOI:10.1080/21580103.2018.1482794.

Meragiaw Misganaw. (2017). Role of Agroforestry and Plantation on Climate Change

Mitigation and Carbon Sequestration in Ethiopia. Journal of Tree Sciences, 36(1), 1.
https://doi.org/10.5958/2455-7129.2017.00001.2.

Mesfin Abebe & Wubalem Tadesse. (2003). Eucalyptus in Ethiopia: risk or opportunity In
M. A. & W. Tadese (Ed.), Ethiopian Institute of Agricultural Research, Ethiopia (Issue
November). January,2014 ,pp 65.

Missanjo, E., Kamanga-thole, G., & Bonongwe, D. (2015). Allometric Equations for
Estimation of Above Ground Biomass of Eucalyptus camaldulensis in Malawi .Journal of
Basic and Applied Research International , 2(2) 2395 - 3438.

Mohd Zaki, N. A., Latif, Z. A., & Suratman, M. N. (2018). Modelling above-ground live trees
biomass and carbon stock estimation of tropical lowland Dipterocarp forest: integration
of field-based and remotely sensed estimates. International Journal of Remote Sensing,
39(8), 2312–2340. https://doi.org/10.1080/01431161.2017.1421793

Montagu, K. D., Düttmer, K., Barton, C. V. M., & Cowie, A. L. (2005). Developing general
allometric relationships for regional estimates of carbon sequestration - An example
using Eucalyptus pilularis from seven contrasting sites. Forest Ecology and Management,
204(1), 115–129. https://doi.org/10.1016/j.foreco.2004.09.003

Moussa, M., Abasse, T., Abdou, I. K., & Larwanou, M. (2019). Applying an indirect method
for estimating and modelling the aboveground biomass and carbon for wood energy in
the arid ecosystems of A ϊ r Tenéré of Niger. 11, 115–126.
https://doi.org/10.5897/JENE2019.0795

Mugasha, W.A., Eid, T., Bollandsås, O.M., Malimbwi, R.E., Chamshama, S.A.O., Zahabu, E.
2012 Allometric models for prediction of aboveground biomass of single trees in miombo
 woodlands in Tanzania. In Proceedings of the First Climate Change Impacts, Mitigation
and Adaptation Programme Scientific Conference. Climate Change Impacts, Adaptation
and Mitigation (CCIAM), Tanzania, pp. 8–18

Mugasha, W. A., Bollandsås, O. M., & Eid, T. (2013). Relationships between diameter and
height of trees in natural tropical forest in Tanzania. Southern Forests, 75(4), 221–237.
https://doi.org/10.2989/20702620.2013.824672.

Mugasha, W. A., Mauya, E. W., Njana, A. M., Karlsson, K., Malimbwi, R. E., & Ernest, S.
(2019). Height-Diameter Allometry for Tree Species in Tanzania Mainland. International
Journal of Forestry Research, 2019. https://doi.org/10.1155/2019/4832849

Mukuralinda, A., Kuyah, S., Ruzibiza, M., Ndoli, A., Leon, N., & Muthuri, C. (2021).
Allometric equations , wood density and partitioning of aboveground biomass in the
arboretum of Ruhande , Rwanda. 3(September 2020).
https://doi.org/10.1016/j.tfp.2020.100050.

Mulugeta Zewdie. (2008). Temporal Changes of Biomass Production , Soil Properties and
Ground Flora in Eucalyptus globulus Plantations in the Central Highlands of Ethiopia.
Doctoral Thesis, Swedish University of Agricultural Sciences Uppsala,74pp.

Mulugeta Zewdie, M., Olsson, M., & Verwijst, T. (2009). Above-ground biomass production
and allometric relations of Eucalyptus globulus Labill . coppice plantations along a
chronosequence in the central highlands of Ethiopia. Biomass and Bioenergy, 33(3):
421–428. https://doi.org/10.1016/j.biombioe.2008.08.007

Návar, J. (2009). Biomass component equations for Latin American species and groups of
species. Annals of Forest Science, 66(2), 208–208.

Nambiar, E. K.Sadanandan.(2021). Small forest growers in tropical landscapes should be

embraced as partners for Green-growth: Increase wood supply, restore land, reduce
poverty, and mitigate climate change.trees Forests and People. 6: 100154.

Nawir, Kassa, H., Sandewall, M., Dore, D., Campbell, B., Ohlsson, B., & Bekele, M. (2007).
Stimulating smallholder tree planting – lessons from Africa and Asia. Unasylva. 58, 53–
 58.

Nelson, B. W., Mesquita, R., Pereira, J. L. G., De, S. G. A., Teixeira, G., & Bovino, L. (1999).
Allometric regressions for improved estimate of secondary forest biomass in the central
Amazon. Forest Ecology and Management .117: 149–167.

Nugroho, N. P. (2014). Developing Site-Specific Allometric Equations for Above-Ground

Biomass Estimation in Peat Swamp Forests of Rokan Hilir District, Riau Province
Indonesia.JournalofForestryResearch,1(1),47–65.
https://doi.org/10.20886/ijfr.2014.1.1.47-65.

Obidzinski, A. K., & Dermawan, A. (2010). Smallholder timber plantation development in

Indonesia : what is preventing progress ? International forestry Rev. 12(4), 339–348.

Ostadhashemi, R., Shahraji, T. R., Roehle, H., & Limaei, S. M. (2014). Estimation of biomass
and carbon storage of tree plantations in northern Iran. Journal of Forest Science, 60(9),
363–371. https://doi.org/10.17221/55/2014-jfs

Otukei, J. R., & Emanuel, M. (2015). Estimation and mapping of above ground biomass and
carbon of Bwindi impenetrable National Park .ALOS PALSAR .4(1), 1–13.

Packard GC (2013) Is logarithmic transformation necessary in allometry? Biol J Lin Soc

109:476–486.

Paulo, J. A., Tomé, J., & Tomé, M. (2011). Nonlinear fixed and random generalized height-
diameter models for Portuguese cork oak stands. Annals of Forest Science, 68(2), 295–
309. https://doi.org/10.1007/s13595-011-0041-y.

Picard, N., L. Saint-André, and M. Henry. 2012. Manual for Building Tree Volume and
Biomass Allometric Equations: From Field Measurements to Prediction,. Rome: Food
and Agricultural Organization of the United Nations, 213 pp

Picard, N., Rutishauser, E., Ploton, P., Ngomanda, A., & Henry, M. (2015). Should tree
biomass allometry be restricted to power models? Forest Ecology and Management, 353,
156–163.
Pohjonen, V. and Pukkala, T., 1990. Eucalyptus globulus in Ethiopian forestry. For. Ecol.
Manage.,36: 19-31.

Poorter, L., Bongers, L., & Bongers, F. (2006). Architecture of 54 Moist-Forest Tree Species :
Traits , Trade-Offs , and Functional Groups Accessed : 07-06-2016 19 : 87(5), 1289–
1301.

Pravalprukskul, P. (2018). Assessing the contribution of Grevillea robusta trees to farm-scale

carbon fluxes in the Central Highlands of Kenya. Master thesis, University of
Copenhagen,Kneya,74pp

Rahman, M. S., Donoghue, D. N. M., Bracken, L. J., & Mahmood, H. (2021). Biomass
estimation in mangrove forests: A comparison of allometric models incorporating species
and structural information. Environmental Research Letters, 16(12).

Ravindranath N. H., Madelene Ostwald.(2008). Carbon Inventory Methods Handbook for

Greenhouse Gas Inventory, Carbon Mitigation and Roundwood Production Projects.
ISBN-13: 978-1-4020-6546-0.

Ribeiro Sabina Cerruto, Carlos Pedro Boechat Soares, Lutz Fehrmann4 , Laércio Antônio
Gonçalves Jacovine Klaus von Gadow (2015). Above and below ground biomass and
carbon estimates for Aboveground and belowground biomass and carbon estimates for
clonal eucalyptus trees in southeast brazil.39(2): 353-363

Rocha, S. M. G., Vidaurre, G. B., Pezzopane, J. E. M., Almeida, M. N. F., Carneiro, R. L.,
Campoe, O. C., Scolforo, H. F., Alvares, C. A., Neves, J. C. L., Xavier, A. C., & Figura,
M. A. (2020). Influence of climatic variations on production, biomass and density of
wood in eucalyptus clones of different species. Forest Ecology and Management,
473(82), 118290.

Roshetko, James M.Bertomeu, Manuel (2015). Multi-species and multifunctional smallholder

tree farming systems in Southeast Asia: timber, NTFPs, plus environmental benefits,
Annals of Silvicultural Research,39(02): 62-69.

Rutishauser, E., Noor’an, F., Laumonier, Y., Halperin, J., Rufi’ie, Hergoualch, K., & Verchot,
 L. (2013). Generic allometric models including height best estimate forest biomass and
carbon stocks in Indonesia. Forest Ecology and Management, 307, 219–225.
https://doi.org/10.1016/j.foreco.2013.07.013

Sawadogo, L., Savadogo, P., Tiveau, D., Dayamba, S. D., Zida, D., Nouvellet, Y., Oden, P. C.,
& Guinko, S. (2010). Allometric prediction of above-ground biomass of eleven woody
tree species in the Sudanian savanna-woodland of West Africa. Journal of Forestry
Research, 21(4), 475–481. https://doi.org/10.1007/s11676-010-0101-4

Segura, M. (2005). Allometric Models for Tree Volume and Total Aboveground Biomass in a
Tropical Humid Forest in Costa Rica 1. 37(1), 2–8.

Shamaki, S., Akindele, S., Isah, A., & Mohammed, I. (2016). Height-diameter Relationship
Models for Teak (Tectona grandis) Plantation in Nimbia Forest Reserve, Nigeria. Asian
Journal of Environment & Ecology, 1(1):1–7. https://doi.org/10.9734/ajee/2016/3063

Stegen James C., Nathan G. Swenson , Renato Valencia , Brian J. Enquist , Jill Thompson .
(2009). Global Ecol. Biogeogr. 18 : 617–625

Sturrock, R. N., Frankel, S. J., Brown, A. V, Hennon, P. E., Kliejunas, J. T., Lewis, K. J.,
Worrall, J. J., & Woods, A. J. (2011). Climate change and forest diseases. Plant
Pathology, 60(1):133–149.

Sullivan, M. J. P., Lewis, S. L., Hubau, W., Qie, L., Baker, T. R., Banin, L. F., Chave, J.,
Cuni‐Sanchez, A., Feldpausch, T. R., & Lopez‐Gonzalez, G. (2018). Field methods for
sampling tree height for tropical forest biomass estimation. Methods in Ecology and
Evolution, 9(5):1179–1189.

Tegegne Derbe, Sisay Yehuala, and Agitew Genanew. (2018). Factors Influencing
Smallholder Farmers Adoption of Eucalyptus Woodlot in Wogera District , North Gondar
Zone , (IJSRM),06(07): EM-2018-566-574. https://doi.org/10.18535/ijsrm/v6i7.em07.

Toirambe, B., Kearsley, E., Haulleville, T. De, Hufkens, K., Baert, G., Huygens, D., Kebede,
Y., Defourny, P., Bogaert, J., & Beeckman, H. (2013). Conventional tree height–diameter
relationships significantly overestimate aboveground carbon stocks in the Central Congo
 Basin Elizabeth. Nature Communications, 4, 2269. https://doi.org/10.1038/ncomms326

Tsegaye Bekele. (2015). Integrated Utilization of Eucalyptus globulus grown on the Ethiopian
Highlands and its Contribution to Rural Livelihood : A Case Study of Oromia , Amhara
and Southern Nations Nationalities and People ’ s Regional State Ethiopia. International
Journal of Basic and Applied Sciences, 4(2), 80–87.

Uller Heitor Felippe, Laio Zimermann Oliveira, Aline Renata Klitzke, Jackson Roberto
Eleotério , Alfredo Celso Fantini, Alexander Christian Vibrans.(2019). Southern
Forests.81(3): 2070-2620. https://doi.org/10.2989/00306525.2019.1581498

Urban, J., Holušová, K., Menšík, L., Čermák, J., & Kantor, P. (2013). Tree allometry of
Douglas fir and Norway spruce on a nutrient-poor and a nutrient-rich site. Trees Structure
and Function. 27(1) : 97–110. https://doi.org/10.1007/s00468-012-0771-y.

Vashum KT, J. (2012). Methods to estimate above ground biomass and carbon stock in natural
forests a review 2157 7625 . J Ecosyst Ecogr ISSN:2157-7625 JEE, an Open Access
Journal, 2(4.1000116), 2 of 8. https://doi.org/10.4172/2157-7625.1000116

Vieilledent, G., Vaudry, R., Andriamanohisoa, S. F. D., Rakotonarivo, O. S., Randrianasolo,

H. Z., Razafindrabe, H. N., Rakotoarivony, C. B., Ebeling, J., & Rasamoelina, M. (2012).
A universal approach to estimate biomass and carbon stock in tropical forests using
generic allometric models. Ecological Applications, 22(2), 572–583.

Vieira, S. A., Alves, L. F., Aidar, M., Araújo, L. S., Baker, T., Batista, J. L. F., Campos, M.
C., Camargo, P. B., Chave, J., Delitti, W. B. C., Higuchi, N., Honorio, E., Joly, C. A.,
Keller, M., Martinelli, L. A., De Mattos, E. A., Metzker, T., Phillips, O., Dos Santos, F.
A. M. (2008). Estimation of biomass and carbon stocks: The case of the Atlantic Forest.
Biota Neotropica, 8(2), 21–29. https://doi.org/10.1590/S1676-06032008000200001

Williams. Richard J., Ayalsew Zerihun Kelvin D. Montagu, Madonna Hoffman, Lindsay B.
Hutley and Xiaoyong Chen .(2005). Australian Journal of Botany,53; 607-619

Xiang, W., Zhou, J., Ouyang, S., & Zhang, S. (2016). Species-specific and general allometric
equations for estimating tree biomass components of subtropical forests in southern
 China. European Journal of Forest Research, 135(5), 963–979.
https://doi.org/10.1007/s10342-016-0987-2.

Xu, Y., Zhang, J., Franklin, S. B., Liang, J., Ding, P., Luo, Y., Lu, Z., Bao, D., & Jiang, M.
(2015). Improving allometry models to estimate the above- and belowground biomass of
subtropical forest, China. Ecosphere, 6(12), art289. https://doi.org/10.1890/ES15-00198.1

Yehualashet Belete, Abere Fentahun, Birhanu Kebede, Teshome Soromessa.(2019).

Nondestructive allometric model to estimate aboveground biomass: an alternative
approach to generic pan-tropical models. Journal of Environmental Studies and Sciences,
hal-02265451. https://hal.science/hal-02265451.

Yordanos sete & Getahun Abreham. (2021). Estimating Land Expectation Value of
Smallholder Farmers ’ Eucalyptus Woodlot Products in Wogera. Reasearch Sqaure.

Zhang Gong,Sangram Ganguly , Ramakrishna R. Nemani , Michael A. White , Cristina

Milesi , Hirofumi Hashimoto , Weile Wang ,SassanSaatchi , Yifan Yu , Ranga B.
Myneni.(2014). Remote Sensing of Environment.1-13.

Zianis, D., & Mencuccini, M. (2005). Aboveground net primary productivity of a beech
(Fagus moesiaca) forest: A case study of Naousa forest, northern Greece. Tree
Physiology, 25(6), 713–722. https://doi.org/10.1093/treephys/25.6.713.

APPENDIX
Appendix 1: Figure 1: Photo taken during field data collection
Appendix 1: Figure 2:The photos show laboratory analysis of samples taken in the Laboratory

BILOGRAPHY

The author was born in central Gondar, Gondar zone at the specific place of college kebele 18
on November 08/1990 E.C. She attends her elementary class at kebele 16 (Leuale Alemayhu)
elementary school and then her higher class attended sheneta secondary school. After the
completion of her high school, she joined her BSc degree program at ambo University College
of Agriculture and veterinary science in the department of Forestry She is gold modelist from
female students of College of Agriculture and veterinary.After graduated, she joined the
Bahirdar University in 2019 to pursue her MSc program in environment and climate change
by Bahirdar university Talent scholarship. Now she is working in Gondar university colleges
of agriculture and environmental science Forestry department.