Biomass and carbon sequestration prediction models for acacia mangium willd plantations in Thai Nguyen province, Viet Nam

UNIVERSITY OF THE PHILIPPINES LOS BAÑOS

Doctor of Philosophy in Forestry: Forest Resources Management

HUNG TUAN NGUYEN

BIOMASS AND CARBON SEQUESTRATION PREDICTION

MODELS FOR Acacia mangium Willd PLANTATIONS

IN THAI NGUYEN PROVINCE,VIET NAM

TEODORO R. VILLANUEVA, Ph.D.

Adviser

Date: ___________________________

This dissertation can be made available to the general public YES

This dissertation can be accessed only after consultation with the

author and dissertation adviser ---------

This dissertation can be accessed only by those bound by confidentiality

agreement ---------

.

__________________________________

HUNG TUAN NGUYEN

__________________________________

TEODORO R. VILLANUEVA, Ph.D.

BIOMASS AND CARBON SEQUESTRATION PREDICTION

MODELS FOR Acacia mangium Willd PLANTATIONS

IN THAI NGUYEN PROVINCE, VIET NAM

HUNG TUAN NGUYEN

JULY 2018

The dissertation attached here to, entitled “BIOMASS AND CARBON

SEQUESTRATION PREDICTION MODELS FOR Acacia mangium

Willd PLANTATIONS IN THAI NGUYEN PROVINCE, VIET NAM”

prepared and submitted by NGUYEN TUAN HUNG in partial fulfillment of the

requirements for the degree of DOCTOR OF PHILOSOPHY (FORESTRY: FOREST

RESOURCES MANAGEMENT) is hereby accepted.

WILFREDO M. CARANDANG

Member, Advisory Committee

__________________________

Date Signed

JUAN M. PULHIN

Member, Advisory Committee

__________________________

Date Signed

MYRNA G. CARANDANG

Member, Advisory Committee

__________________________

Date Signed

TEODORO R. VILLANUEVA

Chair, Advisory Committee

__________________________

Date Signed

Accepted in partial fulfillment of the requirements for the degree of DOCTOR OF

PHILOSOPHY (FORESTRY: FOREST RESOURCES MANAGEMENT).

MARGARET M. CALDERON

Director, Institute of Renewable Natural Resources

__________________________

Date Signed

JOSE V. CAMACHO, JR

Dean, Graduate School

University of the Philippines Los Baños

__________________________

Date Signed

iii

BIOGRAPHICAL SKETCH

The author was born on March 29, 1980 in Thai Nguyen City, Thai Nguyen

Province, Vietnam. He is the eldest of two children of Mr Nguyen Van Hoi and Mrs

Nguyen Thi Nhan. He finished his elementary, secondary, and high school education

from Thai Son, Quang Trung, and Luong Ngoc Quyen School, respectively in Thai

Nguyen City, Thai Nguyen Province, Vietnam in 1998.

He completed his bachelor‟s degree in Forestry from the Thai Nguyen University

of Agriculture and Forestry (TUAF), Thai Nguyen City, Thai Nguyen Province in 2002.

Through the Australia Development Scholarship (ADS), he was able to earn his Master‟s

degree in Forest Science and Management at the Southern Cross University, New South

Wales, Australia in 2012.

In 2015, the Southeast Asian Regional Center for Graduate Study and Research in

Agriculture (SEARCA) granted him a scholarship to pursue his PhD degree in Forestry:

Forest Resources Management at the College of Forestry and Natural Resources,

University of the Philippines Los Banos (UPLB).

He is currently employed as a teacher and researcher at the Faculty of Forestry,

Thai Nguyen University of Agriculture and Forestry, Thai Nguyen Province, Vietnam.

He is happily married to Ms. Pham Thi Hoai, with whom he is blessed with a son,

Nguyen Lam Khoa and a lovely daughter, Nguyen Tue An.

HUNG TUAN NGUYEN

iv

ACKNOWLEDGMENT

First and foremost, my sincerest thanks to my institution, Thai Nguyen University

of Agriculture and Forestry for allowing me to go on study leave and to the Southeast

Asian Regional Center for Graduate Study and Research in Agriculture (SEARCA) for

awarding me a scholarship to pursue a PhD degree in UPLB.

Special thanks are extended to Dr. Teodoro R. Villanueva, Chair, advisory

committee, for his intellectual and professional guidance, critical comments,

encouragement and remarkable interest in supervising this study. My grateful

acknowledgment also to the members of my advisory committee: Dr Myrna G.

Carandang, Dr Wilfredo M. Carangdang, and Dr Juan M. Pulhin for their valuable

comments, sincere concern, and understanding.

I am also thankful to the rest of the faculty and administrative staff of the Institute

of Renewable and Natural Resources and the Faculty of UPLB Graduate School for their

great support.

A note of gratitude also goes to all the members of the Faculty of Forestry, Thai

Nguyen University of Agriculture and Forestry for the support and assistance extended to

me for my study. My deepest gratitude goes to my loving family for all their sacrifices and

encouragements. I am forever grateful to my loving wife, Mrs. Pham Thi Hoai, my son,

Mr. Nguyen Lam Khoa and and my daughter Ms Nguyen Tue An for their love and

spiritual support.

Special thanks also go to all of my sincere friends in Vietnam, as well as in UPLB

who have directly or indirectly helped me during my stay at UPLB and also during the

conduct of dissertation research in my country.

v

TABLE OF CONTENTS

CHAPTER PAGE

Title Page i

Approval Page ii

Biographical Sketch iii

Acknowledgement iv

Table of Contents v

List of Tables ix

List of Figures xi

Abstract xiv

I INTRODUCTION 1

Background of the Study 1

Statement of the Problem 8

Hypothesis of the Study 12

Objective of the Study 13

Importance of the Study 14

Scope and Limitation of the Study 17

II REVIEW OF LITERATURE 18

Biomass and Carbon Sequestration of Forest Ecosystems 18

Studies on Biomass and Carbon Sequestration in the World 23

Studies on Biomass and Carbon Sequestration in Vietnam 27

Studies on Biomass and Carbon Sequestration 31

Studies on Biomass and Carbon Sequestration of Acacia Species

in the World

31

vi

CHAPTER PAGE

Studies on Biomass and Carbon Sequestration of Acacia

species in Vietnam

33

Biomass and Carbon Prediction Techniques 36

Aboveground Biomass 37

Below Ground Biomass (BGB) 39

Biomass Growth Models in Forest Management 41

Estimation of Total Aboveground Biomass and Carbon Storage

based on IPCC Method

50

Estimation of Total Aboveground Biomass and Carbon Storage

based on the Jenkins et al. (2003) Method

50

Estimation of Total Aboveground Biomass and Carbon Storage

based on the Chojnacky and Jenkins (2010) Method

51

III METHODOLOGY 55

Study Area 55

Data Gathering Techniques 58

Analytical Framework 59

Mapping of Study Site 62

Sample Size and Sampling Method 63

Tree Volume 65

Basal Area of Stand 66

Stand Density 67

Mean of Diameter 68

Mean Total Height 68

Aboveground Biomass (AGB) Calculation from Volume 69

Biomass and Carbon Estimation of Individual Trees 70

Individual Trees Biomass Estimation 70

vii

CHAPTER PAGE

Methods to Determine theAmount of Carbon Stored in the

Plantation

75

Data Analysis and Model Development 77

Model Application and Validation 85

IV RESULTS AND DISCUSSION 87

Acacia Plantation Management in Thai Nguyen Province 87

Description of the Study Area 91

Descriptive Statistics for All Variables of the Study 92

Stand Variables 92

Sample Variables 98

Biomass and Carbon Estimation of Sample Tree and Stand 104

Green Biomass of Individual Tree and Stands 105

Dry Biomass and Carbon of Individual Trees and Stands 116

Percentage between Dry and Green Biomass for Sample Tree by

Age

122

Carbon and Carbon Dioxide Stock of Individual Trees and

Stands

123

Total Aboveground and Belowground Biomass Estimation 128

Model Development, Selection and Evaluation for Biomass Estimation 131

Tree Fresh Biomass Models by Separated Ages 134

Bole Biomass Models by Separated Ages 144

Biomass Models of Branch and Leaves Combination by

Separated Ages

151

Green Biomass Models for all Age Levels 159

Dry Biomass Models by Separated Ages 161

viii

CHAPTER PAGE

Dry Biomass Models for All Age Levels 169

Carbon Prediction Models by Separated Ages 171

Carbon Estimation Models for All Age Levels 178

Carbon and Carbon Dioxide Stock for Present Land Use and Scenarios

Plantation Management in Thai Nguyen Province

180

Carbon and Carbon Dioxide Stock of Present Land Use of

Acacia mangium Willd

180

Scenario 1. Land Use Change 181

Scenario 2: Trend in biomass and carbon of Acacia mangium

Willd Plantation

182

V SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS 186

Summary 186

Conclusion 188

Recommendation 191

LITTERATURE CITED 194

ix

LIST OF TABLES

TABLE PAGE

1 Estimated belowground biomass and carbon models for evergreen forests in

Central Highlands of Vietnam.

41

2 The estimated biomass models built for the forest types over the world. 52

3 The estimated above ground biomass models built for forest types in

Vietnam.

53

4 Inventory data sheet for each sample plot. 65

5 Green biomass calculated from the sample plots. 72

6 Dry biomass calculated from the sample plots 74

7 Forest resources of Thai Nguyen Province. 89

8 Descriptive summary of stand variables of Acacia mangium Willd. 93

9 ANOVA test of diameter and tree height at different strata conditions. 86

10 Descriptive statistics for sample trees of Acacia mangium Willd by ages. 101

11 Average green biomass distribution in the tree (kg/tree). 106

12 Dry biomass of sample trees by age. 118

13 Proportion of dry weight over green weight by age. 122

14 Average carbon and carbon dioxide of sample trees and components. 124

15 Aboveground and belowground biomass estimation from existing equations 130

16 Biomass model parameters and their performance criteria for Whole-trees. 135

17 Selection of best models for whole-trees biomass estimation by age. 136

18 Actual and predicted biomass for whole-tree by age. 140

19 Bole biomass model parameters and their performance criteria for different

age.

145

20 Selected models for bole biomass estimation by ages for Acacia mangium

Willd.

146

x

TABLE PAGE

21 Comparison between actual and predicted bole biomass for the best

equations.

147

22 Models for combined branch and leaves of Acacia mangium Willd by ages

of plantation.

153

23 Selected models for biomass prediction of branch and leaves combined. 154

24 Actual and predicted values of biomass for branch and leave combined

(kg/tree).

155

25 Green biomass models tested for all age classes of Acacia mangium Willd.

Equation in bold font was selected as best model for biomass prediction.

160

26 Total dry biomass equation tested with predicted variables at different age. 163

27 The best selected dry biomass equations with predicted variables by age. 164

28 Actual and predicted dry biomass by age. 165

29 Dry biomass model tested for all age classes of Acacia mangium Willd. The

equation in bold font indicates the best models for dry biomass prediction.

170

30 Tree carbon prediction models, parameters, and their performance criteria. 173

31 The best selected equations for carbon prediction by age. 174

32 Comparison of actual and predicted carbon sequestration of whole-tree by

age.

175

33 Carbon model tested for all age classes of Acacia mangium Willd. Model 4

(D*H) in bold font, was selected as the best model for carbon prediction.

179

34 Current total carbon sequestration and three scenarios that happened. 182

35 Biomass, carbon, and carbon dioxide trend values 183

xi

LIST OF FIGURE

FIGURE PAGE

1 Estimated US atmospheric CO2 mitigation requirements and potential

sequestration capacities (Sundquist et al., 2008).

2

2 The global carbon cycle. Fluxes shown are approximate for the period

2000-2005, as reported by the IPCC (Sundquist et al., 2008).

20

3 Principal global carbon pools. Soil organic carbon (SOC) and soil inorganic

carbon (SIC) are two different forms of carbon that can be stored in the soil

(Lal 2004a).

20

4 Five carbon pools in the forest ecosystem. 22

5 Circle plot used in America. 38

6 Map of Thai Nguyen Province. 57

7 Analytical framework of Acacia mangium Willd biomass and carbon

model.

61

8 Temporary sample plot of Acacia mangium Willd plantation. 63

9 Sample plots sketch out in the Acacia mangium Willd plantation. 64

10 Sample trees felled and weighed for green biomass in the field. 71

11 Samples for dry biomass were calculated from green biomass. 73

12 Maps showing location of the study site. 92

13 Average diameter and total height of Acacia mangium Willd at different

strata conditions by age.

97

14 Sample tree variables of Acacia mangium Willd by age. 103

15 Green biomass (kg/tree) changed by ages of sample trees at the different

strata (A, B, C, D). Average at bottom, side, and top biomass of the strata,

respectively).

108

16 Biomass distributions in the tree components by ages (%). 110

17 Average of green biomass by sample tree and stand by age. 113

18 Correlation analysis of plantation age and green biomass. 114

xii

FIGURE PAGE

19 Correlation analysis of relationship of total green biomass and diameter and

height.

115

20 Dry biomass (kg/tree) changed by ages of sample trees at the different

strata (A, B, C, D are average, bottom, side and top biomass of the strata

respectively).

119

21 Dry biomass distribution in sample tree components (%) by age. 120

22 Average dry biomass of Acacia mangium Willd sample trees (kg/tree) and

stand (ton/ha) in Thai Nguyen Province.

121

23 Correlation analysis of log-transformation of D, H and dry weight of

Acacia mangium Willd sample trees in all age levels.

121

24 Average carbon and carbon dioxide of sample trees and stands. 125

25 Correlation analysis of Acacia mangium Willd plantation age and carbon

stock.

128

26 Actual versus predicted whole-tree biomass using the best equations for

separated different ages. The R2 value indicates the goodness of fit for

fitted regression line

142

27 Actual and predicted whole-tree biomass using the best equations for all

age levels. The R2 value indicates the goodness of fit for fitted regression

line.

145

28 Relationship between actual and predicted biomass of sample tree bole

(kg/tree) by age.

150

29 Average actual vs predicted biomass of sample trees for every year of

plantation.

151

30 Tested actual and predicted biomass by ages using linear regression scatter. 158

31 Residual analysis plotted of all sample trees for the best selected models 161

32 Correlation analysis between measured and predicted dry biomass by age. 168

33 Correlation analysis between whole-tree dry biomass and diameter, total

height.

169

34 Residual analysis plots for all dry biomass of sample trees for the best

selected model.

170

35 Correlation analysis between actual and predicted carbon stock by age. 177

xiii

FIGURE PAGE

36 Correlation between whole-tree carbon with diameter and total height of

sample.

178

37 Residual plotted analysis of all sample trees for the best selected models. 180

38 The trend of biomass and carbon of Acacia mangium Willd in Thai

Nguyen.

184

xiv

ABSTRACT

Hung Tuan Nguyen, University of the Philippines Los Banos, July 2018. Biomass and

Carbon Sequestration Prediction Models for Acacia Mangium Willd Plantations in

Thai Nguyen Province, Vietnam.

Major Professor: Dr Teodoro R. Villanueva

The study developed a model to estimate current biomass and carbon stocks as well as

predict future biomass and carbon sequestration potential for forest plantations of Acacia

mangium Willd in Thai Nguyen Province, Vietnam. Specifically, the study: 1) characterized the

Acacia mangium Willd plantation in Thai Nguyen Province, Vietnam; 2) estimated the current

biomass and carbon stocks of tree and stand for Acacia mangium Willd plantations; 3) developed

a biomass and carbon models for tree of Acacia mangium Willd; 4) determined the future

conditions of plantation based on the programs and policies of the government; and 5)

recommended appropriate management strategies to improve the forest plantation development

and management.

A total of 126 plots representing various ages of plantations were established at the

bottom, hillside, and hilltop of the plantation. Data collected from each plot included age of

plantation, spacing, density, diameter, total height, basal area, and volume. Estimates of the

various plantation characteristics showed significantly higher values in the bottom compared with

those in the other parts of the plantations sampled.

The data for biomass and carbon estimation and development of prediction model came

from 54 destructive sampled trees of different diameter classes (big, medium, and small) of the

different ages. Six candidate non-linear regression equations using variables as diameter, total

height, and age of plantation were tested and assessed for statistical validity and accuracy in

biomass and carbon prediction. Data analysis was carried out in Excel and STATA 14 PM

software. The study showed that the major biomass and carbon of tree are boles, followed by

branches and leaves. Biomass and carbon models were tested for separated ages (each age class

was tested by the model), as well as all age levels from ages 2 to 7 Acacia mangium Willd

plantation (all age classes from 2 to 7 were tested by the model). In terms of separated ages, the

model with one variable as diameter (D) showed the better values than variable height (H) and

two variables (D, H) combined due to the high correlation efficiency (R2

), small standard error

(SE), and higher F values. As for the models tested for all age levels combined with the addition

of the variable age (A), there was no significant difference observed between single predictor and

combined predictors. The accuracy of the values was tested by chi-square and residual analysis to

compare between observed and predicted biomass and carbon.

The prediction equations were used to assess future biomass and carbon sequestration in

the province. Scenarios of biomass and carbon change were assessed based on the programs and

policy of the government.