UTILIZATION OF OPTICAL SATELLITE DATA FOR MEASURING
CARBON DIOXIDE AT FELDA MAOKIL
MUHAMMAD NOOR HAZWAN BIN JUSOH
A project report submitted in partial fulfillment of the
requirement for the award of the degree of Master of
Engineering (Civil - Environmental Management)
Faculty of Civil Engineering
Universiti Teknologi Malaysia
NOVEMBER 2009
iii
To my beloved mother and father
JUSOH JAAFAR
KAMARIAH CHE SOH
&
My loving brothers and sisters
MUHAMMAD NOOR HISYAM
MUHAMMAD NOR HANIF
NUR MADIHAH
NUR NADIAH SABIHAH
MUHAMMAD NOOR BAIHAQI
NUR HABIBAH SAFIAH
MUHAMMAD NOOR SUFFIAN
iv
ACKNOWLEDGEMENTS
“In the Mighty Name of Allah, The Most Beneficent, The Most Merciful”
Firstly, I would like to express my sincere gratitude and love to my dear parents
Jusoh Jaafar and Kamariah Che Soh and also my cherished brothers and sisters for the
consistent support, guidance and encouragement during my study.
Secondly, a lot of thanks to my supervisor Dr. Mohd Badruddin Mohd Yusof for
his attention and advice in order to ensure this thesis are completed. Not to forget sincere
appreciation to my co-supervisor, PM. Dr. Ab. Latif Ibrahim from whom I received the
necessary guidance throughout my final project. Thank you so much to both of you.
Lastly, I express very thankful to my friends who had shared ideas and
knowledge that contributed on my writing. I also wish to thank all staffs of
Environmental Engineering Laboratory, Universiti Teknologi Malaysia; Pak Usop, Mr.
Azrin and Mr. Suhaimi because of their passionate and helpful I’m manage to make this
Master’s project much easier.
Thank you so much for all of you which I’m not mention here. Your contribution
directly or indirectly I always keep in my heart.
Thank you…
v
ABSTRACT
Currently there are many developments especially in construction’s sectors
where a lot of air pollutants have been produced and resulted in the degradation of the
environment. The issue has grown in importance in light of recent global warming.
This study is based on a short term observation of carbon dioxide (CO2) released due to
a road construction at Felda Maokil in Segamat. The method that has currently been
used to determine Net Primary Productivity (NPP) at global scale using satellite data
known as Moderate Resolution Imaging Radiometer (MODIS). The purpose of this
study is to measure the amount carbon dioxide (CO2) that may be absorbed by oil palm
tree for an area of about 450 hectares in Oil Palm plantation with low resolution MODIS
satellite data. Several objectives have been carried out such as to determine the amount
of Net Primary Productivity (NPP) in Felda Maokil, to assess the accuracy of MODIS
data by validation process with in-situ data and lastly, to assess the correlation of NPP
value from 2001, 2005 and 2009. The data were obtained through two sources which
are in-situ as primary source and satellite data as secondary source. Data from four year
interval had been chosen to insure more variations during the study period. Therefore
the study was verified by the NPP values for year 2001, 2005 and 2009. The resulted
acquired presented the mean NPP values for the three years processed were 468.169 g
Cm2/y (2001), 560.685 g Cm2/y (2005) and 541.781 g Cm2/y (2009). Meanwhile,
Normalized Difference Vegetation Index (NDVI) analysis had gave the resulted as 0.696
(2001), 0.863 (2005) and 0.716 (2009). As a conclusion, the mean NPP values for three
years of MODIS images processed have shown similarities and road development had
gave some impacts on these results.
vi
ABSTRAK
Pada masa kini terdapat banyak aktiviti pembangunan yang melibatkan sektor
pembinaan telah menyebabkan pembebasan udara tercemar ke atmosfera. Fenomena
pencemaran ini secara tidak langsung telah membawa kepada berlakunya pemanasan
global. Oleh itu, kajian ini dijalankan di Felda Maokil, Segamat bagi menentukan
kandungan karbon dioksida (CO2) yang disebabkan oleh pembinaan jalan. Kaedah yang
digunapakai adalah data satelit yang turut dikenali sebagai Moderate Resolution Imaging
Radiometer (MODIS).
Selain itu, pengukuran CO2 dapat diketahui bilangannya
berpandukan terma Net Primary Productivity (NPP).
Tujuan kajian ini dijalankan
adalah untuk mendapatkan kandungan CO2 yang mampu diserap oleh pokok kelapa
sawit di dalam kawasan ladang kelapa sawit berkeluasan 450 hektar.
Kajian ini
mengandungi beberapa objektif iaitu menentukan jumlah kandungan NPP di Felda
Maokil, menilai ketepatan data MODIS dengan melakukan proses pengesahan data dari
tapak kajian dan menilai hubungan bagi data NPP untuk tahun 2001, 2005 dan 2009.
Terdapat dua kaedah dalam mendapatkan data kajian iaitu persampelan data di tapak
kajian (sumber utama) dan data satelit ( sumber kedua). Perbezaan 4 tahun bagi tahun
kajian dipilih supaya banyak perubahan dapat dilihat bagi tempoh kajian dijalankan.
Keputusan kajian akan disahkan berdasarkan nilai NPP bagi tahun 2001, 2005 dan 2009.
NPP bagi tiga tahun data proses ialah 468.169 g C m2/y (2001), 560.685 g C m2/y
(2005) dan 541.781 g Cm2/y
(2009).
Sementara itu, analisis untuk Normalized
Difference Vegetation Index (NDVI) memberikan
keputusan 0.696 (2001), 0.863
(2005) dan 0.716 (2009). Sebagai kesimpulan, nilai purata NPP telah menunjukkan
persamaan untuk tiga tahun data MODIS. Disamping itu, aktiviti pembinaan jalan turut
memberi impak terhadap keputusan kajian.
vii
TABLE OF CONTENT
CHAPTER
TITLE
TITLE PAGE
DECLARATION
DEDICATION
ACKNOWLEDGEMENT
ABSTRACT
ABSTRAK
TABLE OF CONTENTS
LIST OF TABLES
LIST OF FIGURES
LIST OF SYMBOLS
LIST OF ABBREVIATIONS
LIST OF APPENDICES
PAGE
i
ii
iii
iv
v
vi
vii
x
xi
xiv
xv
xiv
CHAPTER
1
1.1
1.2
1.3
1.4
1.5
INTRODUCTION
General
Problem Statement
Aim and Objectives
Scope of the Study
Expected Finding
1
1
3
4
4
5
CHAPTER
2
2.1
2.2
2.3
LITERATURE REVIEW
Carbon Sequestration
Carbon Cycle
Sources of Carbon Dioxide
Transportation
Industrial
Generating Electricity
6
6
7
7
8
9
10
2.3.1
2.3.2
2.3.3
viii
2.4
2.5
2.6
2.7
2.8
2.9
2.9.1
2.9.2
2.10
2.11
2.12
2.13
2.14
2.15
2.16
2.16.1
2.16.2
2.16.3
2.17
2.18
2.19
2.20
2.20.1
2.20.2
2.21
CHAPTER
3
3.1
3.2
3.3
3.4
3.5
3.6
3.7
Carbon in Atmosphere
Atmospheric Increase of Carbon Dioxide
Carbon Pollutants
Pollutant Materials
Pollutant Deposition
Types of Gasses in The Atmosphere
Carbon Monoxide
Nitrogen Oxides
Photosynthesis
Functioning of Photosynthesis Process
Classification of Ecosystem Types
The Basis of Net Primary Productivity
Potential Net Primary Productivity
Annual Assessment of NPP Allocation
Limiting Factors
Shortage of Nutrients
Some Climatic Factors
Carbon Dioxide
Malaysia’s Legislation
Kyoto Protocol
MODIS Satellite Data
Carbon Dioxide Capture and
Geologic Storage
Techniques of Carbon Dioxide Capture
Geological Storage of Carbon Dioxide
Previous Research on Related Studies
Using MODIS Satellite Data
12
12
13
13
14
14
15
15
16
17
18
20
21
21
22
23
23
26
26
27
30
METHODOLOGY
Introduction
Study Area
Data Collection
Carbon Dioxide Meter
Algorithm
Spectral Bands
ERDAS Software
41
41
41
42
43
44
46
47
31
33
36
38
ix
CHAPTER
4
4.1
4.2
4.3
4.4
4.5
4.5.1
4.5.2
4.5.3
4.5.4
4.6
CHAPTER
5
5.1
5.2
RESULTS AND DISCUSSIONS
Image for Propose Location from
MODIS Data
The Amount of Net Primary Productivity
And Normalized Difference
Vegetation Index
Correlation between Net Primary
and Normalized Difference Vegetation
values (2001, 2005 and 2009)
Relationship between of APAR and
MODPRI at Study Area
Relationship between Field Measurement
and Meteorological Data
Reliability Test
Student’s t-test
Analysis of Variance (ANOVA)
Pearson’s Correlation Coefficient
Effects of Road Construction in
Carbon Absorption
CONCLUSION AND
RECOMMENDATIONS
Conclusion
Recommendations
49
49
53
62
64
69
69
71
71
72
74
80
80
81
REFFERENCES
84
APPENDICES
87
x
LIST OF TABLES
TABLE NO
TITLE
Table 2.1
U.S. Carbon Dioxide Emissions from Transportation Sector
Energy Consumption
Emissions which Produced by Industrial Processes
Amount of Atmospheric Carbon Monoxide
for Primary Sources
Greenhouse Gas Emissions for Some Countries from
the year 1992 to 2007
Change in Greenhouse Gas Emissions (1990 to 2004)
The Amount of Annual Carbon Dioxide Emissions
from Major Industrial Sources
MODIS – Related Research Topics
Net Primary Production Values at Felda Maokil
in unit of g C m2/d
Normalized Difference Vegetation Index Values
at Felda Maokil in unitless
APAR values at Felda Maokil
MODPRI values at Felda Maokil
Net Primary Production values at tolerable range
Normalized Difference Vegetation Index
values at tolerable range
ANOVA for NPP
ANOVA for NDVI
Correlation between NPP and NDVI
Table 2.2
Table 2.3
Table 2.4
Table 2.5
Table 2.6
Table 2.7
Table 4.1
Table 4.2
Table 4.3
Table 4.4
Table 4.5
Table 4.6
Table 4.7
Table 4.8
Table 4.9
PAGE
8
10
16
30
31
35
37
55
55
67
67
72
72
74
74
75
xi
LIST OF FIGURES
FIGURE NO
TITLE
Figure 2.1
The Carbon Dioxide Emissions from Industrial Activities
and Land Use Change
Schematic Map of the Major Biomass of the World
Annual NPP and Annual Precipitation
The Length of the growing season (P) in the months, The
Average Temperature (T) during the growing season in
°C, and The Day Length (D) in hours as a function of
latitude
The Relationship between Dry Matter Production (P)
and Total Transpiration (W) of Oats Grown in Containers
The Flow of Carbon Dioxide According to the Concept
of Carbon Dioxide Capture and Geological Storage
Carbon Dioxide Capture and Storage
The Processes that occur from Carbon Dioxide
Capture Techniques
The Geological Storage Options for Carbon Dioxide
The Project Planning for the Road Constructions that
shorten the trip between Chaah and Bukit Kepong
Carbon Dioxide Meter Model 7515
The Lists of Spectral Bands to Produce MODIS
Satellite Image
Work Schedule
MODIS Satellite Image for Year 2001
MODIS Satellite Image for Year 2005
MODIS Satellite Image for Year 2009
Figure 2.2
Figure 2.3
Figure 2.4
Figure 2.5
Figure 2.6
Figure 2.7
Figure 2.8
Figure 2.9
Figure 3.1
Figure 3.2
Figure 3.3
Figure 3.4
Figure 4.1
Figure 4.2
Figure 4.3
PAGE
11
20
21
26
27
34
37
37
39
44
46
49
50
52
53
54
xii
Figure 4.4
Figure 4.5
Figure 4.6
Figure 4.7
Figure 4.8
Figure 4.9
Figure 4.10
Figure 4.11
Figure 4.12
Figure 4.13
Figure 4.14
Figure 4.15
Figure 4.16
Figure 4.17
Figure 4.18
Figure 4.19
Figure 4.20
Figure 4.21
Figure 4.22
Figure 4.23
Figure 4.24
Standard Deviation for NPP Value at Felda Maokil
(2001, 2005 and 2009)
Standard Deviation for NDVI Value at Felda Maokil
(2001, 2005 and 2009)
The Different Colour of Net Primary Production (2001)
The Different Colour of Net Primary Production (2005)
The Different Colour of Net Primary Production (2009)
The Different Colour of Normalized Difference
Vegetation Index (2001)
The Different Colour of Normalized Difference
Vegetation Index (2005)
The Different Colour of Normalized Difference
Vegetation Index (2009)
Scatter Plots of NPP and NDVI values
in year 2001 at Felda Maokil
Scatter Plots of NPP and NDVI values
in year 2005 at Felda Maokil
Scatter Plots of NPP and NDVI values
in year 2009 at Felda Maokil
A Graph of APAR values at different years
A Graph of MODPRI values at different years
Relationship between APAR and MODPRI
at Felda Maokil (2001)
Relationship between APAR and MODPRI
at Felda Maokil (2005)
Relationship between APAR and MODPRI
at Felda Maokil (2009)
The Range of Area Covered in A Normal Distribution
Location of Road Construction at Study Area
(2001) based on NPP
Location of Road Construction at Study Area
(2005) based on NPP
Location of Road Construction at Study Area
(2009) based on NPP
Location of Road Construction at Study Area
(2001) based on NDVI
56
56
58
59
60
61
62
63
64
65
65
67
68
69
69
70
72
78
79
80
81
xiii
Figure 4.25
Figure 4.26
Location of Road Construction at Study Area
(2005) based on NDVI
Location of Road Construction at Study Area
(2009) based on NDVI
82
83
xiv
LIST OF SYMBOLS
CO2
-
Carbon Dioxide
CFC
-
Chlorofluorocarbons
CH4
-
Methane
K
-
Potassium
N2O
-
Nitrus Oxide
N
-
Nitrogen
P
-
Phosphorus
xv
LIST OF ABBREVIATIONS
CCS
-
Carbon Dioxide Capture and Geologic Storage
DOE
-
Department of Environmental
EOS
-
Earth Observing System
EOR
-
Enhanced Oil Recovery
ECBMR
-
Enhanced Coal Bed Methane Recovery
GHG
-
Greenhouse Gas
GPP
-
Gross Photosynthetic Products Produced
IEA
-
International Energy Agency
LUC
-
Light Use Efficiency
MODIS
-
Moderate-Resolution Imaging Spectroradiometer
MRSO
-
Malaysian Rectified Skew Orthomorpic
NPP
-
Net Primary Production
SBRS
-
Santa Barbara Remote Sensing
UNFCCC
-
UN Framework Convention on Climate Change
WEC
-
World Energy Council
xvi
LIST OF APPENDICES
APPENDIX
A
B
C
TITLE
Manual Calculation for Reliability Test
Manual Calculation for Student’s t-test
Manual Calculation for Analysis of
Variance (ANOVA)
PAGE
90
92
94
CHAPTER 1
INTRODUCTION
1.1
General
Recently, human activities have contributed more critical problems to the
environment. There are a lot of environmental issues that arise since last decade
such as water quality, solid waste, air pollution, noise pollution, etc. Besides that,
the climate change that occurred has some side effects from this matter.
As example, acid rain where emissions of sulphur oxides, nitrogen oxides and
hydrocarbons are transformed in the atmosphere into sulphate and nitrate particles.
Combination of sunlight and water vapour will produce a complex chemical reaction
in mild sulphuric or nitric acid. This acid rain will occurs when pH levels falls below
5.6. Another effect that arise is global warming where solar radiation that come to
earth is not fully release from atmosphere and the results it increase the earth
temperature since more radiation is absorbed by the earth’s surface; melting the ice at
Antarctic and Antarctica which lead to increase water level. The situation happen
where more greenhouse gas; carbon dioxide (CO2), chloroflorocarbons (CFC),
methane (CH4), nitrus oxide (N2O) and tropospheric ozone in atmosphere
2
In order to maintain the quality of environment and prevent before it become
worst, some affords has been taken out in order to cater this environment issues.
Solid waste can be control by several measures such as open dumping, landfill,
incinerator, transfer stations and shrinking waste stream. However, this prevention
cannot be successful as there are some factors affecting waste generation such as
location, seasons, eating habits/lifestyle, etc. Meanwhile, for noise pollution can be
control by management during construction phase, planning on construction sites
activities, prevention from design stage of machine and additional control after
machine is built. Lastly is air pollution which comes from natural sources, stationary
and vehicles. Usually, air pollutants can be dividing into several types which are
particulate (PM10), nitrogen oxide (NOx), sulphur oxide (SOx), carbon (COx) and
hydrocarbon (HC). Since stationary produces more pollution to atmosphere, it can
be cater by using cyclone separator, baghouse filter, electrostatic precipitator and wet
scrubber.
In the past few years, the global MODIS (Moderate Resolution Imaging
Radiometer) NPP (Net Primary Productivity) has been measured using small scale
on-ground flux tower measurements using the eddy covariance method. So far this
method is limited to implement due to a number of flux tower worldwide. Until now
only about 450 sites flux tower distributed worldwide. Based on previous research
by Cohen et al., (2003) and Turner et al., (2005) several issues have arise regarding
the appropriateness of the variety of product scale such as to match the low spatial
resolution of MODIS satellite data with plot scale flux tower measurements on the
ground.
3
Vargas et al., (2007) and Falgae et al., (2001) have mentioned that although
there a lot of flux tower such as AsiaFlux in Asia, KoFlux in Korea, OzFlux in
Australia and others seems insufficient for the validation of the global MODIS NPP,
thus cannot produced an accurate measurement of global NPP. Besides that, MODIS
NPP comes in various models such as initiated by MOD17 (Running et al., 1999),
“continuous field model” (Rahman et al., 2004) and “Carnegie-Ames-Stanfor
Approach model” (CASA) (Jinguo et al., 2006).
1.2
Problem Statement
Over the past century there has been a dramatic increase in the amount of
gases that releases from several sources have contributes air pollution to the
environment. So far, however, there has been little discussion about greenhouse
effect that increases the earth temperature.
In Malaysia basically, awareness among publics about air pollution are still
lacking since not too much campaign and activities that related to this environmental
issues. Besides that, poor of monitoring from government agencies have make the
quality of gases that releasing are higher than allowable in air quality standard. At
construction site especially, most of air quality at this locations are not taken
seriously. Enforcement will take place only when there are complains make by local
people or it become serious issues.
4
As mention earlier, the global warming is the process whereby the earth’s
temperature is getting increases.
The most primary factor that cause of global
warming is carbon dioxide emissions.
This significantly occurs due too many
sources which producing this gas such as natural activity, power plants, cars, trucks,
aircrafts and others sources which not mention here. According to the report, 8
billion tons of CO2 have been release into atmosphere in year 2008. In records, 40%
of all CO2 emissions are cause by power plants, 33% of all the CO2 sent forth is the
product of cars and trucks and 3.5% are released from aircraft. Therefore, based on
this facts it show that CO2 gases playing major role in increasing temperature which
known as global warming.
Therefore, plants play in either sequestering atmospheric carbon or releasing
carbon into the atmosphere. This study is conducted to measure carbon in the
atmosphere that will absorb by plants. Since Malaysia is located at rain forest
region, the amount of CO2 should be low.
1.3
Aim and Objectives
This study is conducted in order to assess the efficiency of MODIS data using
ERDAS software. Therefore, to achieve the aim several objectives are needed as a
guideline;
i.
To determine the amount of Net Primary Productivity (NPP) at road
construction in Felda Maokil, Segamat.
ii.
To assess the accuracy of MODIS data by validation process with in-situ
data.
iii.
To assess the correlation of NPP value for years; 2001, 2005 and 2009.
5
1.4
Scope of the Study
Main focus in this study is involve the use MODIS satellite data at propose
location. The implementation for this data is decided because it easy to obtain by
downloads from website at http://ladsweb.nascom.nasa.gov. Besides that, MODIS
data also is free to get and since the MODIS views the entire surface of the earth
every one to two days, the current data is available.
The MODIS satellite will be analyze using ERDAS IMAGE V9.1 from Leica
Geosystems Geospatial Imaging. The use of this software because of it is practical to
apply and well known product.
Besides that, the propose location for this study is situated at Felda Maokil in
Segamat, Johor Darul Takzim. Several points will be allocated in order to measure
CO2 concentration. This parameter is observed by using special equipments like
Carbon Dioxide (CO2) Meter for gaseous concentration.
Then, CO2 concentration data will be assessing using ERDAS IMAGE V9.1
from Leica Geosystems Geospatial Imaging. An Addition, these values between
MODIS satellite data and insitu data will be compare in order to measure the quality
of both data.
6
1.5
Expected Finding
The study would be expected to find that the NPP values are low since the
surrounding area was palm oil trees. Besides that, the propose location is less
developed which lack gasses released into the atmosphere. Although there are must
be some error between MODIS satellite data and insitu data, the differences between
both data are in small amounts. The NPP value for MODIS satellite data is more
accurate than in-situ data. This is because of the influence by surrounding areas such
as wind faster, respiration rates by human and animals, activities by local people and
others.
CHAPTER 2
LITERATURE REVIEW
2.1
Carbon Sequestration
Based on previous research by Sakaki (2009), forests can play a role in either
sequestering atmospheric carbon or releasing carbon into the atmosphere. In another
word, carbon sequestration can be define as a process where atmospheric carbon dioxide
is absorbed by trees through photosynthesis and store as carbon in biomass (trunks,
branches, foliage and roots) and soils. Moreover, the ability of forest to sequester
additional carbon can be increase by sustain the forestry practice and this action also
helps in enhancing ecosystem services; improves soil and water quality. However,
human activities such as harvesting and regenerating forests will give a result in net
carbon sequestration in wood products and new forest growth. In order to increase
forest carbon, this problem can be preventing by restoring forested ecosystems, planting
trees and improving forest health.
7
2.2
Carbon Cycle
Previous research has reported by Bolin (1986) that he has presented estimates of
the major carbon pools and flux rates. This is because pool sizes, transfer patterns and
flux rates of the carbon cycle are not fully or very accurately appreciated. In our
atmosphere consist of several matters which are 720 x 109 tons of carbon, the biota 830
x 109 tons and below ground pools include 60 x 109, 500 x 109, 1400 x 109 and 5000 x
109 tons in surface detritus, peat, soil and fossil fuels respectively. This situation,
however, the natural net input of carbon dioxide to the atmosphere from vegetative
systems is close to zero where the system in undisturbed states. Besides that, the
disturbance by forest fires, organic matter decomposition and plant respiration will be
balanced by carbon dioxide uptake for photosynthesis process in existing and newly
established vegetation. This statement however, being argued by Lugo and Brown
(1986) where frequent and minor disturbances to forests are common and that forests
slowly sequester atmospheric carbon.
2.3
Sources of Carbon Dioxide
The arising amounts of carbon dioxide because of human activity mainly and
some of them are coming natural activity. These sources are giving significant effects to
the atmosphere such as greenhouse effects, smog, respiration problems and others.
Therefore, identification of carbon sources can help in minimizes the amount of CO2
gases that produces.
8
2.3.1
Transportation
Currently, the increasing numbers of vehicles commonly generate a lot of CO2
that releases to the atmosphere. There are many strategies that promoted by government
as the way to control CO2 emissions such as car pooling, wider use of public transport
and regular inspection and maintenance of vehicles, yet it’s still not fully effective in
overcome the issues. Petroleum combustion is the largest source of carbon dioxide
emissions in the transportation sector, as opposed to electricity-related emissions in the
other end-use sectors. In a large country especially, the numbers of transportation have
give more impacts to the environment through the releasing of CO2 by human activity.
Table 2.1: U.S. Carbon Dioxide Emissions from Transportation Sector Energy
Consumption
(Source: Emissions of Greenhouse Gases in the United States 2006, 2007).
9
2.3.2 Industrial
Industrial sector is the major elements that releasing abundant of CO2 gases to
the atmosphere.
Through their manufacture processes, without any monitoring by
Department of Environmental (DOE) the emissions will create the serious problems to
human being at surrounding area. Normally, the amount of CO2 that produces are
depending on the raw materials itself and how it been processing. According to the
report which conducted in year 2004, industrial processes generated emissions of 320.7
teragrams of CO2 equivalent, or 5% of total U.S. greenhouse gas emissions. Besides
that, also in year 2004 the record shows that the CO2 emissions from all industrial
processes were 152.6 Tg CO2 Eq which mean it was 3% of national CO2 emissions. The
figure below shows the amounts of gasses that produces in industrial sector differ by
manufacture process.
10
Table 2.2: Emissions which Produced by Industrial Processes.
(Sources: Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2004)
2.3.3 Generating Electricity
Based on previous research by Chris et al. (2002), he found that there are three
major complications which associating CO2 emissions with electricity consumption.
According to the practice, electricity usually can be generated from different primary
sources and most of them will create CO2 emissions (e.g., coal combustion) while minor
result in virtually no CO2 emissions (e.g., hydro).
11
At the second place, the mix of generation resources used to meet loads may vary
at different times of day or in different seasons. Lastly, the generation sources related to
electricity usage can be difficult to trace due to transportation of electrical energy over
long distances by complex transmission and distribution systems.
Besides that, the emissions may occur far from jurisdiction in which hat energy
is consumed. Therefore, it can conclude that the emissions resulting from electricity
consumption vary considerably on when and where it is used since this affects the
generation sources providing the power.
(Source: Climate Change information kit, UNEPIUC, 1997)
Figure 2.1: The CO2 Emissions from Industrial Activities and Land Use Change.
12
2.4
Carbon in Atmosphere
CO2 is the gas that produces by photosynthesis processes. Richard and Steven,
(1998) mentioned that half the gross photosynthetic products produced (GPP) are
expended by plants in autotrophic respiration (Ra) for the synthesis and maintenance of
living cells, releasing CO2 back into atmosphere. The remaining carbon products (GPPRa) go into net primary production (NPP); foliage, branches, stems, roots and plant
productive organs. Other than that, atmospheric carbon dioxide concentrations and the
availability of soil nitrogen (N) must also need to be considered when modelling
photosynthesis, carbon allocation and respiration.
2.5
Atmospheric Increase of Carbon Dioxide
In the middle of nineteenth century, the concentration of carbon dioxide in the
global concentration has been estimated to have 270 + 10 ppm (4.8 x 105 μ g/m3).
However, this amount increased with the carbon concentration of approximately 350
ppm (6.3 x 105 μ g/m3).
Additionally, the rate of increasing carbon dioxide
concentration is in the amount of 1.5 ppm (2.7 x 105 μ g/m3) annually. Trabalka et al.,
(1986) points out that if this condition continuously happen, the carbon dioxide amount
in the global atmosphere may be nearly two times of present value.
13
2.6
Carbon Pollutants
There are two major sources that contribute to atmospheric contaminants,
namely, carbon monoxides and carbon dioxides. Both pollutants are very important
atmospheric contaminants where carbon monoxides consist of potent mammalian
toxicity while carbon dioxide has ability to regulate global temperature. Based on U.S.
Environmental Protection Agency (1976), carbon monoxide has not been shown to
produce acute effects on plants at concentrations below 100 ppm (11.5 x 104 μg m3) for
exposures from 1 to 3 weeks. Besides that, the hypothesis shows that the increasing
concentration of carbon dioxide in the atmosphere will result in elevated global
temperatures.
2.7
Pollutant Materials
According to William (1939), air pollutants can be defined as materials that
occur in the troposphere in quantities in excess of normal amounts. Usually the air
pollutants that founded are in solid, liquid and gaseous conditions and these characters
are depending on both natural and human anthropogenic processes. Air pollution that
occurs based on natural sources are volcanic and other geothermal eruptions, forests
fires, gases released from vegetation, wind-blown soil and other debris, pollen, spores
and sea spray particles.
Other than that, air pollution that causes by human
anthropogenic processes includes a variety of combustion and industrial activities. Air
pollutants can be divided into two terms which is primary pollutants and secondary
pollutants.
14
Pollutants that can be categorizes as primary pollutants where particulate or
gaseous materials released directly into the troposphere in large amounts by natural or
anthropogenic processes. Meanwhile for secondary pollutants, particulate or gaseous
materials formed in the atmosphere from precursors released in large amounts to the
troposphere by natural or anthropogenic processes.
2.8
Pollutant Deposition
The air pollutant can be transferred by a variety of deposition mechanisms and
these mechanisms usually can be divided into wet and dry deposition. The application
of wet deposition is related with the movement of dissolved gasses and large particles;
diameter greater than 20 μm, via incident precipitation. This process will lead to occur
of acid precipitation which is known as rain or snow that having pH less than 5.6.
Normally, the water that contain of pH values below will becomes acid then above 7
will becomes alkaline. Previous studies have reported by Lindberg et al., (1982) have
shown that pollutant concentrations in rain generally decrease as rainfall amount,
intensity and duration decrease but amount of rain recorded, however, exerts the most
significant influence on pollutant transfer.
15
2.9
Types of Gasses in the Atmosphere
The amounts of gasses towards air pollution are increasing in the atmosphere.
This situation occurs based on the current condition of human activities. Besides that,
there are also several sources that are well known to contribute to the air pollution such
as vehicles, burning, industrial area and others.
2.9.1 Carbon Monoxide
The concentration of carbon monoxide is different with carbon dioxide where
this gas appears to be stable or only slightly increasing in clean atmospheres remote
from excessive local input of carbon oxide. The existing of carbon monoxide in the
atmosphere is in very approximately equal amounts from anthropogenic sources and
from the biota on a global basis as shown in Table 2.3. Although, the concentration of
carbon monoxide is not increasing dramatically, despite increasing combustion of fossil
fuels, yet assuming the process of carbon monoxide is occurring.
16
Table 2.3: The Amount of Atmospheric Carbon Monoxide
Source
Methane oxidation
Anthropogenic
Biota
Ocean
Terrestrial plants
Chlorophyll degradation
Min
400
600
100
20
300
CO (106 tons yr -1)
Total
Northern Hemisphere
Max
Min
Max
4000
200
2000
1000
540
900
220
40
90
200
14
140
700
200
500
(Source: Nozhevnikova and Yurganov, 1978)
2.9.2 Nitrogen Oxides
The releasing of nitric oxide and nitrogen dioxide to the atmosphere are coming
from the high temperature combustion of fossil fuels for transportation, energy
generation and the manufacture of petroleum products.
In their previous studies,
Soderlund and Svensson (1976) have made estimated that 19 x 1012 g of nitrogen as the
annual anthropogenic input via nitrogen oxide release to the atmosphere. This amount
however, might be increase in the future where the utilizing of higher combustion
temperature will increase the atmospheric input of nitrogen oxides from fossil fuel
burning. Besides that, there are also some natural mechanisms that helping on releasing
nitrogen oxides to the atmosphere such as fixation by lighting (Junge, 1958; Ferguson
and Libby, 1971), inflow from stratosphere (Soderlund and Svensson, 1976), chemical
conversion from ammonia in the troposphere (Crutzen, 1974; McConnell, 1973) and loss
of gaseous nitric oxide from soils (Robinson and Robbins, 1975).
17
2.10
Photosynthesis
Photosynthesis can be defined as the process that plants convert atmospheric CO2
to carbon products.
This process takes place within cells containing chloroplasts.
Chloroplasts contain chlorophyll and other pigments that absorb sunlight. The chemical
reactions below shows the energy from the sun causes electrons to become excited and
water molecules where it is split into hydrogen and oxygen.
ʹ
൅ ൅ ื ൅ ................................................................ (2.1)
ʹ
ʹ
ʹ
Photosynthesis is restricted by both physical and biochemical processes and
involves some reactions that require light and others that can take place in the dark. At
the leaf surface, stomata limit the diffusion of carbon dioxide into intercellular spaces.
While inside leaves, CO2 must dissolve in water and pass through cell walls to reach the
sites where chemical reactions take place within chloroplast. In the day light reactions,
radiation absorbed by chlorophyll causes excitation of electrons which are transferred
down to the chain of specialized pigment molecules to reaction centres where highenergy compounds are formed, water is split and O2 released as shown in chemical
reactions above. Meanwhile, in the dark reactions, plants use the enzyme ribulosebisphosphate carboxylase-oxygenase (Rubisco) for the primary fixation of CO2. When
the light comes, photorespiration also occurs in the process of generating the substrate
ribulose bisphosphate (RuBP), normally when the ratio of O2 and CO2 increases in the
chloroplast (B. Berg and C. McClaugherty, 2008).
18
2.11
Functioning of Photosynthesis Process
There are many types of components that influence of living organic matter such
as C, H and O and then followed by N and P, accounting for 95% of plants matter and a
number of other elements in minor proportions that are important to various
physiological functions of the plants. Based on Redfield et al. (1963), he found that the
ratio which called as Redfield ratio where it’s come from the production of organic
matter by freshwater and marine planktonic organisms takes carbon, nitrogen and
phosphorus in the atomic ratio C:N:P = 106:16:1. In addition, this ratio also applied for
the phytoplankton and the photosynthetic reaction between CO2 and aqueous nitrate and
phosphate ions, the chemical reaction as shown below;
ͳͲ͸
̴
൅
ʹെ
൅ ͳ͸͵ ൅ ൅ ͳͺ ൅ ͳʹʹ ֎
Ͷ
ʹ
ʹ
ቀ ቁ
ቀ ቁ ቀ ቁ ൅ ͳ͵ͺ .................................................... (2.2)
ͳͲ͸
͵ ͳ͸
͵
ʹ
Ͷ
ʹ
Referring to the reaction above, respiration or oxidation of organic matter,
produces CO2, nitrate, phosphate and water where thus the reaction proceed from right
to the left. Meanwhile an under anaerobic conditions, the reaction that occur will lead to
the formation of CO2, methane and ammonia by heterotrophic bacteria;
ቀ ቁ
ቀ ቁ ቀ ቁ ൅ ͳͶ ͵ ͳ͸
͵
ʹ ͳͲ͸
Ͷ
ʹ
̴
൅
ʹെ
՜ ͷ͵ ൅ ͷ͵ ൅ ͳ͸ ൅ ൅ ͳͶ ......................................... (2.3)
Ͷ
Ͷ
ʹ
Ͷ
19
Several researches have showed that, land plants have an average C:N:P atomic
ratios that vary from 510:4:1 (Delwiche and Likens, 1977) to 822:9:1 (Deevey, 1973)
and 2057:17:1 (Likens et al., 1981).
Compared between land plant and aquatic
phytomass, photosynthesis process by land plant produces organic matter with a
relatively much higher concentration of carbon and this can be prove by reaction below
as example;
̴
൅
ʹ െ
ʹͲͷ ൅ ͳ͹͵ ൅ ൅ ͳͻ ʹͲ͹Ͷ ൌ
ʹ
Ͷ
ʹ
ቀ ቁ
ቀ ቁ ቀ ቁ ൅ ʹͲͻͳ ................................................ (2.4)
͵ ͳ͹
͵
ʹ ʹͲͷ͹
Ͷ
ʹ
In order for photosynthesis to occur in terrestrial and aquatic plants, this process
involves net primary of carbon from surrounding medium into the plant cell. In term of
conditions for terrestrial plants, the medium is the atmosphere with its gaseous CO2,
while for aquatic plants the medium is the water with its dissolved inorganic carbon.
2.12
Classification of Ecosystem Types
Based on study by Lieth and Whittaker, (1975) the earth surface consists of
different kinds of vegetation associated with different environmental features. All of
these differences can be expressed in net primary productivity variations which are of
vital importance for the self maintenance or management of the respective ecosystems.
The prime importance for the optimal use of individual types of ecosystem is the
knowledge and understanding of variation that consists in NPP and their causes.
20
There are some difficulties on classification of ecosystem types because of
incompleteness of the available information, mistakes in classifying local ecosystem
terminologies.
Ovington, (1965) identifies that ‘woodland’ may be applied to a
vegetation which lacks a continuous tree canopy but the total vegetation coverage is
continuous where it is also can be applied as a general term for forest. However, the
terms ‘forest’ is used for real closed forests as well as for more open woodlands.
According to Rubel, (1930); Ellenberg and Muller-Dombois, (1967); Schmithusen,
(1968); Schmidt, (1969); Walter, (1973) and UNESCO, (1973) stated that the design of
various classification systems are depending on different criteria and remain to some
degree subjective.
The terms that have been used such as ‘vegetation unit’ or
‘ecosystem type’ are applied to any grouping of plants and are not limited. Therefore,
they are the perfectly safe terms to use to designate a band of tropical forest or march
vegetation.
(Source: After Odum, 1971)
Figure 2.2: Schematic Map of the Major Biomass of the World
21
2.13
The Basis of Net Primary Productivity
The formation of glucose that come out of water and carbon dioxide within under
the action of light is the elements that exists in the production of organic plant materials.
According to previous research Goudriaan and Ajtay (1979), water vapour inevitably
escapes when the plant takes up carbon dioxide from the air and this situation usually
occurs in terrestrial plants. Therefore, when the plant loses too much water and is
threatened by drought, it particularly reacting by closing the stomatal openings where
both water loss and CO2 assimilation are reduced. Because of that, water may be a
limiting factor for net primary productivity.
(Source: Whittaker, 1970)
Figure 2.3: Annual NPP and Annual Precipitation
22
2.14
Potential Net Primary Productivity
By assuming variables to be optimal can help in measuring the potential net
primary productivity (NPP). There is standard procedure to take the climatic conditions
for given external parameters and only to assume an optimal water and nutrient supply.
However, the changing of climatic conditions is very limited for man’s ability. Besides
that, at the certain conditions such as incoming radiation, suboptimal temperatures and
the length of the vegetation period are still limiting the crop production. The experiment
by Loomis and Gerakis (1975) for the potential yields shows that by providing water and
nutrients optimally available and in disease-free cultures.
Referring to natural
vegetations, since one or more factors are suboptimal the circumstances are not
favourable. Therefore, the most fruitful way to analyze deviations from the potential
productivity is to consider one factor as limiting.
2.15
Annual Assessment of NPP Allocation
Annual primary production represents all carbon sequestered into dry matter
during a year and is equivalent to total carbon uptake through photosynthesis minus the
loss through autotrophic respiration. According to the available practice, net primary
productivity (NPP) is estimated by assuming the growth of all tissue produced during a
year, whether or not the tissue was consumed by herbivores or entered the detrital pool.
The statement above can be shown below.
NPP = B + DB + CB.................................................................................................... (2.5)
23
From that equation, B is the change in biomass over a period of a year, DB is
detritus produced during a year and CB represents consumption of biomass by herbivores
during the year.
Eis et al., (1965) and Pregitzer and Burton (1991) claim that in particular
climatic zone, the annual pattern of carbon allocation to foliage and stem wood shows a
general consistency, except when large crops of seeds are produced or unusual weather
conditions prevail.
According to destructive analysis of trees, information can be
obtained on how growth is distributed. Meanwhile Whittaker and Woodwell, (1968);
Kira and Ogawa, (1971); Gholz et al., (1979); Dean, (1981); Pastor et al., (1984)
mention that proportional increments in biomass of stem wood, leaves, branches and
large diameter roots are related exponentially to increases in stem diameter. Production
is determined by periodic measurement of stem diameter or by extracting wood cores
and measuring annual increments.
A previous study by Sukwong et al., (1971) found that variable-plot surveys are
generally more efficient because only a few trees require measurement at each sampling
point. Therefore biomass increment is calculated by measuring all trees within a known
area or by using variable-plot surveys based on the diameter of trees intercepted by a
selected angle.
24
2.16
Limiting Factors
Limiting factors are the elements that influence the effectiveness of plants to
sequester the CO2. The factors which identifies are shortage of nutrients, some climatic
factors and carbon dioxide.
2.16.1 Shortage of Nutrients
Part of the nutrients will available be to the plants during microbial
decomposition and this situation happen when harvesting is not implementing where
organic materials that produced will return to the soil as fallen leaves, seeds, etc.
Sometimes mature ecosystems for the net primary productivity may approach the
potential NPP. This is because of influencing the additional of the quantities released by
weathering of soil materials and to those supplied by rain and microbial nitrogen
fixation.
Without fertilizer application, the annual yield will reach a rather low
equilibrium level, determined by the natural sources of nutrients. The practice applied in
primitive agriculture activities like shifting cultivation, the stock of nutrients is made
available to a crop by forest burning.
Reasonably the crop is well supplied with
nutrients in the first few years, but the level soon declines through uptake and leaching
(Bolin et al., 1979)
25
Hambridge, (1949) and Natr, (1975) point out that deficiency of an elements not
only shows up in a characteristic symptoms, but it invariably reduces the net primary
production. For the plant nutrients, the most important are nitrogen (N), phosphorus (P)
and potassium (K) but other elements also required in smaller amounts such as Ca, Mg,
Fe, Cu and Mn. Although each nutrients has their own role in specific processes but the
elements which needs to take consideration is nitrogen where this nutrient often limiting
and the most suitable for calculation.
2.16.2 Some Climatic Factors
Due to a decrease in length of the growing season, this condition will affecting
the decline in NPP with decreasing mean annual temperature. Etherington (1975) found
that mean annual temperature and mean annual precipitation can be considered as
factors that determine the type of natural vegetation. The species composition changes
to such an extent that net primary productivity is not limited by mean temperature, since
the value is higher than 10C and it is occurring within changing climatic conditions.
Hence, the influence on net primary production cannot be taken as a whole
representative for the influence of temperature on the performance of a single species.
26
(Source: Larcher, 1973)
Figure 2.4: The Length of the growing season (P) in the months, The Average
Temperature (T) during the growing season in C, and The Day Length (D) in hours as a
function of latitude
Besides that, the actual growing period of the vegetation may be shorter than the
period permitted by temperature because of water shortage. If this situation builds up
gradually, it will contribute more difficult for the plant to withdraw water from the soil.
In order to mitigate the problems, some plants react by closing their stomata. The
difference between two types of plants shows that the plants that close their stomata are
savers while the others are spenders that use fast water. In another major study, Lof
(1976) found that the ratio of transpired water and dry matter formed between these
types of plants are decline with fertilization because of nitrogen increases dry matter
production but not the transpiration rate.
27
(Source: De Wit, 1958)
Figure 2.5: The Relationship between Dry Matter Production (P) and Total
Transpiration (W) of Oats Grown in Containers
2.16.3 Carbon Dioxide
Several studies conducted by Goudriaan and Van Laar (1978); Gifford and
Musgrave (1970) show that even under light saturation, saturation with CO2 still occurs
at 300ppm. Meanwhile in many cases, a CO2 enrichment of the ambient external air
increases the net assimilation rate at light saturation.
28
The difference of saturation effect caused by CO2 induced stomatal closure. The
CO2 concentration in the substomatal cavity is approximately constant for a numbers of
plant species but when the CO2 assimilation rate rises with increasing radiation intensity,
the stomata open up further to compensate the CO2 depletion in the cavity below them.
Previous study has reported that CO2 regulation of stomatal resistance may be
induced by water stress. As additional, in glasshouses CO2 enrichment may not be
extrapolated to field conditions because of CO2 fertilization are not provided which lead
to the decrease of the CO2 concentration in the air until 100ppm. Usually in the field
where there is much better turbulent exchange with the atmosphere, the CO2
concentration is not reduced to less than 250ppm.
2.17
Malaysia’s Legislation
Mostly, the successes in completing projects depend on proper machineries that
required at construction site. This situation however, tends to contribute emission to the
environment since the majority of machineries are using diesel engines which pollute
more than petrol engines. Besides that, at certain site locations, they are involved with
heavy machineries and these situations usually generate more air pollution.
The
machineries condition also needs to be taken into account whereby at construction site,
the maintenance matter is not properly considered. Therefore, this situation will make
the fuel being burnt completely which may result in an increase amount of gasses being
released. Based on these problems, all the contractors and developers in Malaysia are
enforced to follow the guideline as stated in the Environmental Quality (Control
Emission from Diesel Engines) Regulations 1996.
29
2.18
Kyoto Protocol
The name of Kyoto Protocol was given by the location of city where it was
agreed to by negotiators in December 1997, is a treaty intended to implement the
objectives and principles agreed in the 1992 UN Framework Convention on Climate
Change (UNFCCC). This protocol is constructing and establishes in order to binding
commitment for the reduction of four gasses which are carbon dioxide, methane, nitrous
oxide, sulphur hexafluoride (Michael, 2008). An additional, there are also another two
groups of gasses which produced by “Annex I” (industrialized) nations (e.g.,
hydrofluorocarbons and perfluorocarbons). However, the successful of this protocol
required general commitments for all member countries. Kyoto Protocol contains five
principle concepts which include;
i.
Commitments. The heart of the Protocol lies in establishing commitments for
the reduction of greenhouse gases that are legally binding for Annex I countries,
as well as general commitments for all member countries.
ii.
Implementation.
In order to meet the objectives of the Protocol, Annex I
countries are required to prepare policies and measures for the reduction of
greenhouse gases in their respective countries. In addition, they are required to
increase the absorption of these gases and utilize all mechanisms available, such
as joint implementation, the clean development mechanism and emissions
trading, in order to be rewarded with credits that would allow more greenhouse
gas emissions at home.
iii.
Minimizing Impacts on Developing Countries by establishing an adaptation fund
for climate change.
iv.
Accounting, Reporting and Review in order to ensure the integrity of the
Protocol.
v.
Compliance. Establishing a Compliance Committee to enforce compliance with
the commitments under the Protocol.
30
Referring to Kyoto Protocol, the members who come from industrialized
countries are needed to reduce their collective GHG emissions by 5.2% compared to the
year 1990. National limitations range from 8% reductions for the European Union and
some others to 7% for the United States, 6% for Japan, and 0% for Russia. The treaty
permitted GHG emission increases of 8% for Australia and 10% for Iceland. Besides
that, Kyoto Protocol also includes Emissions Trading, the Clean Development
Mechanism and Joint Implementation which can define as flexible mechanisms. The
applications of these mechanisms whereby to allow Annex I economies to meet their
greenhouse gas (GHG) emission limitations by purchasing GHG emission reductions
credits such as through financial exchanges, projects that reduce emissions in nonAnnex I economies, from other Annex I countries or from Annex I countries with excess
allowances. Based on practice that available this means that Non-Annex I economies
have no GHG emission restrictions, but however, it have financial incentives to develop
GHG emission reduction projects to receive carbon credits that can then be sold to
Annex I buyers as encouraging sustainable development.
The differences between
Annex I and non-Annex I where Annex I entities typically will want to acquire carbon
credits as cheaply as possible, while Non-Annex I entities want to maximize the value of
carbon
credits
generated
from
their
domestic
Greenhouse
Gas
Projects
(http:///www.wikipedia.com).
Table 2.4: Greenhouse Gas Emissions for Some Countries from the year 1992 to 2007
Country
India
China
United States
Russian Federation
Japan
Worldwide Total
Change in greenhouse gas Emissions
(1992-2007)
+103%
+150%
+20%
-20%
+11%
+38%
(Source: Kyoto Protocol, http://wikipedia.com)
31
According to the prediction that was made by the Intergovernmental Panel on
Climate Change (IPCC), an average global rise in temperature of 1.4°C (2.5°F) to 5.8°C
(10.4°F) between 1990 and 2100. Since that matter, the Kyoto Protocol has come with
the objective for stabilization of greenhouse gas concentrations in the atmosphere at a
level that would prevent dangerous anthropogenic interference with the climate system.
Table 2.5: Change in Greenhouse Gas Emissions (1990 to 2004).
(Source: Kyoto Protocol, http:///www.wikipedia.com)
32
2.19
MODIS Satellite Data
Several studies have been made involving MODIS satellite data. According to
Chin (2005), Terra lie, the first EOS (Earth Observing System) was launched on
December 18, 1999 which consists of five remote sensing sensors. Among of them, the
most comprehensive EOS sensor is Moderate-Resolution Imaging Spectroradiometer
which also known as MODIS. This technology was provided by NASA where it
managed by NASA’s Goddard Space Flight Centre (GSFC) which is located at
Greenbelt, Maryland. In addition, MODIS was developed by Hughes Corporation’s
Santa Barbara Remote Sensing (SBRS) in Santa Barbara, California.
Syaiful (2008) points out that MODIS consists of 36 bands from 0.405 – 14.385
μm that will collect data at 250 m, 500 m and 1 km spatial resolutions. Besides that
according to Jensen (1996), MODIS contains of unique combination of features where
it’s able to detect a wide spectral range of electromagnetic energy, it takes measurements
at three spatial resolutions, it takes measurements every day and it also has a wide field
of view. Therefore, MODIS enables to collect the data for the entire earth surface at
every 1 to 2 days where observation will be made in 36 co-registered spectral bands at
moderate resolution (0.25 – 1 km) of land and ocean-surface temperature, primary
productivity, land-surface cover, clouds, aerosols, water vapour, temperature profiles
and fires.
33
2.20
Carbon Dioxide Capture and Geologic Storage
Due to the current scenarios of emissions, the International Energy Agency
(IEA), the World Energy Council (WEC), the European Commission or the United
States Department of Energy (DOE) have put the same suggestion that the amount of
energy consumption will rise in between 16 and 18 GtOE by 2030. This result however
is depending on fossil fuels which continuously dominate the energy mix. Therefore, in
order to overcome this issue, Holloway (2007) has stated that the capture and geological
storage (CGS) is technology that can help in reduce the amount of CO2 in atmosphere
from large industrial installations such as fossil fuel-fired power stations by 80-90%. In
CCS processes there are several steps that need to take into accounts which are capture,
transport and storage.
34
(Source: Holloway, 2007)
Figure 2.6: The Flow of Carbon Dioxide According to the Concept of Carbon Dioxide
Capture and Geological Storage
According to the IPCC report for year 2005, fossil-fuelled power generation has
released over 42% of overall anthropogenic CO2 emissions where about 80% of CO2
emissions from the industrial sector. The installation of CO2 capture for conventional
power plants (particularly coal-fired units) and certain other industrial facilities such as
cement mils, refineries, fertilizer factories, steel mills and petrochemical plants seems to
appear to be the most applicable.
35
Table 2.6: The Amount of Annual Carbon Dioxide Emissions from
Major Industrial Sources
In Mt CO2 /yr
Power
10, 539
Iron & Steelmaking
646
Cement Manufacture
932
Oil Refining
798
Petrochemicals
379
Oil & Natural Gas Processing
50
Other Sources (including biomass)
124
Aggregate worldwide large stationary sources of CO2 emissions
13, 468
(Source: IPCC, 2005)
2.20.1 Techniques of Carbon Dioxide Capture
Since the CO2 comes in many different sources, there are three techniques which
will be utilized according to the type of installation. Rubin et al., (2005) has mentioned
that the fossil fuel-fired power plants are the dominants industrial point sources of CO2
in most countries. CO2 may be captured from several techniques by the following;
i.
Post-combustion decarbonisation
This technique is the most mature but among three techniques it was very costly and
is appropriate for existing installation.
Besides that, the post-combustion
decarbonisation involves separating the CO2 contained in combustion gases.
36
ii.
Pre-combustion decarbonisation
By using this technique, the process for this technique consist of treating the fuel
either with steam and air (steam reforming) or with oxygen (partial oxidation) to
produce a synthesis gas that contains mainly carbon monoxide (CO) and hydrogen,
potential energy carrier that generates no CO2 emissions.
Then, the CO will be
converting to the presence of water (H2O) and after that separates the resulting CO2 for
capture and storage.
iii.
Oxyfuel combustion decarbonisation
This technique is still in the pilot phase, but it requires a combustion of gas while is
highly concentrated in CO2 where the volume between 80% and 90%. An addition the
volume of CO2 could constitute a suitable retrofit technology for existing installation.
Besides that, the process for oxyfuel combustion decarbonisation uses high-purity
oxygen instead of air for combustion but the difficulty appears in extracting the oxygen
from the air. The research is still in progress for separation step where the oxygen
supply is derived from a reaction involving a metal oxide. So that, the metal particles
from metal oxide such as iron fillings could be medium in carrying oxygen from air to
fuel.
37
(Source: IPCC, 2005)
Figure 2.7: Carbon Dioxide Capture and Storage
(Source: IPCC, 2005).
Figure 2.8: The Processes that occur from Carbon Dioxide Capture Techniques
38
2.20.2 Geological Storage of Carbon Dioxide
For the purpose of the storage of the CO2 gasses; there are many natural
underground CO2 fields around the world (Studlick et al., 1990; Pearce et al., 1996;
Stevens 2005). Based on practice, at this stage the gas could reach the supercritical state
and thus occupy the smallest possible volume. There are several types of storage that
can be used for geological storage of CO2 such as storage in depleted oil and natural gas
reservoirs, storage in unminable coal beds and storage in saline aquifers (Price et al.,
2008).
i.
Storage in depleted oil and natural gas reservoirs.
The application of this type can help in reducing CO2 emissions since the natural
reservoirs have proven their capacity to contain hydrocarbons for several million years.
Besides that, although this type is not widespread it has been practice among oil and gas
industry where CO2 will be injected into oilfields to reduce crude oil viscosity, improve
mobility and thereby boost the recovery rate and the technique being applied was known
as Enhanced Oil Recovery (EOR). In minimizing the costs, the infrastructure that uses
for exploration and production of crude oil (pipes and wells) can be reused for CO2
storage operations.
ii.
Storage in unminable coal beds.
For the second type of storage, the coal bed is not applied as a reservoir. In this
option, the process for storing the CO2 will be implementing by absorption of the gas.
This technique can be used not only on storage of CO2 but also methane recovery
(ECBMR-Enhanced Coal Bed Methane Recovery), by providing the coal bed is
adequately covered over by impermeable cap rock.
39
iii.
Storage in saline aquifers
The aquifers that are formed of porous, permeable rock often saturated with
brackish water or brine that is unfit to drink are potential storage sites for considerable
quantities of CO2. Besides that, the uses of this type require a sufficient depth which is
more than 800 meters and have overlying impermeable layers.
(Source: Injection and Storage, http://www.co2crc.com.au).
Figure 2.9: The Geological Storage Options for Carbon Dioxide
40
2.21
Previous Research on Related Studies using MODIS Satellite Data
In recent years, the application of MODIS satellite data has been applied in many
difference purposes. A numbers of researchers have used MODIS satellite data in the
study.
Table 2.7: MODIS – Related Research Topics
NO.
AUTHORS
1.
Wahid et al. (2004)
2.
3.
4.
Helen et al. (2006)
Martin et al. (2008)
Latif et al. (2007)
TITLE
Mapping Net Primary in Tropical Rain Forest Using
MODIS Satellite Data
Regional evaporation estimates from flux tower and
MODIS satellite data
Remote estimation of carbon dioxide uptake by a
Mediterranean forest
Evaluation of the MODIS NPP Product for a Japanese
Test Site
Previous study by Wahid et al., (2004) has reported that the estimated values for
distribution of NPP for the whole Peninsular Malaysia and local scale Pasoh Forest
Reserve were 804.37 g Cm2/y and 633.85 g Cm2/ y. Furthermore, the use of Monteith’s
equation as well as micrometeorological approach is being applied in this study to map
local scale NPP in tropical rain forest. Therefore, researchers identifies three types of
MODIS satellite data bands which are utilized to the micrometeorological model which
included visible band, near infra-red band and ocean bands (band 11 and band 12). On
the other hand, MODIS satellite data is obtained from level 1B of morning Terra
MODIS satellite data for Peninsular Malaysia and this requirement have been acquired
from free website at http://ladsweb.nascom.nasa.gov.
41
According to a study by Helen et al., (2006), the aerodynamic resistance-surface
energy balance model and the Penman-Monteith (P-M) equation are two models that
utilized in estimating land surface evaporation at 16-day intervals using MODIS remote
sensing data and surface meteorology as inputs. For this reason, this study investigated
3 years of evaporation and meteorological measurements from evergreen Eucalyptus
forest (a cool temperate) and tropical savanna (a wet/dry) where both these locations are
from two contrasting Australia ecosystems. Consequently, the result showed that the
aerodynamic resistance-surface energy balance approach failed because small error in
the radiative surface temperature translate into large errors in sensible heat, and hence
into estimates of evaporation. In the meantime, the P-M model is sufficiently estimated
the magnitude and seasonal variation in evaporation in both ecosystems (RMSE = 27
Wm - 2, R2 = 0.74), demonstrating the validity of the proposed surface conductance
algorithm. These results however, are not included degradation in the performance of
the P-M when gridded meteorological data at coarser spatial (0.05°) and temporal (daily)
resolution were substituted for locally measured inputs.
In another major study, Martin et al., (2008) claim that the current remote
sensing method for estimating gross primary productivity are not satisfactory. In fact, it
happened because they rely too heavily on three factors which are the availability of
climatic data, the definition of land-use cover and the assumptions of the effects of these
two factors on the radiation-use efficiency of vegetation (RUE). Thus, he suggest that a
new methodology is required in assess RUE and overcome the problems associated with
the capture of fluctuations in carbon absorption in space and over time. In some cases,
the reflectance vegetation indices (e.g. NDVI, EVI) is being used as it allows green plant
biomass and plant photosynthetic capacity to be assessed. Though, the problem rose
since certain vegetation types, for instance the Mediterranean forests, consist of very low
seasonality of these vegetation indices and a high seasonality of carbon uptake whereby
these criteria are important to determine.
42
Besides that, he also points out that there are positive relationships between
photochemical reflectance index (PRI) and RUE which lead to the possibility in
estimated RUE and GPP in real time and then actual carbon uptake of Mediterranean
forests at ecosystem level using the PRI. Finally, it helps to avoid from relying on
untrustworthy maximum RUE.
In recent years, there has been increasing amounts of data from the Moderate
Imaging Radiometer (MODIS) that are being utilized on generate the annual net primary
productivity (NPP) at global scale. In general, the validation process can be achieved by
following two types of procedures such as an accurate ground measurements and
modelling methods. Practically, most of researchers used Eddy covariance flux tower
measurements which utilized a climate model as the way to assess the MODIS NPP
products (Latif et al., 2007). Hence, this study conducts a summation method which one
of the direct on-ground NPP measurements that quoted from a test site in Japan. This
method is then applied to evaluate the MODIS NPP product for the proposed location.
Furthermore, the application of this method also involved on the summation of the tree
growth increment, litter production and grazed amount. Besides that, there is some
equation which is known as Monteith equation that was used for assessing three years
MODIS satellite data (i.e., 2004, 2005 and 2006). By doing both operations (summation
method and the Monteith equation), the comparison between results can be made. As a
result, Latif et al. (2008) concluded that the trend of annual NPP based on two types of
measurements is comparable with an RMSE of 0.8 t/ha yr.
CHAPTER 3
METHODOLOGY
3.1
Introduction
This study is conducted to collect CO2 concentration in atmosphere. The
proposed methodology is applied at study area of Felda Maokil in Segamat, Johor.
The data collection was obtained through two sources; primary and secondary
sources.
At the first stage (primary source) the data was obtained from website by
download at http://ladsweb.nascom.nasa.gov known as MODIS satellite data.
Meanwhile for the second stage (secondary source) the data will be collected using in
situ method. This study will compare between both data using several tests.
4
44
3
3.2
Stud
dy Area
The propose
p
locaation for studdy area is loocated at Feldda Maokil, Segamat.
S
Thhe
d
decision
forr this study area is maade because there is a road constrruction where
s
surrounded
b oil palm plantation annd forests. Therefore
by
T
byy looking at this situatioon,
t study iss conducted to measure the influencce of road coonstruction on the carboon
this
s
sequestration
n by oil palm plantationn. The areaa involves a 22 km roadd constructioon
f
from
existinng one (from
m Felda Maaokil in Segaamat to Bukkit Kepong in Muar) annd
o
only
severall location pooints allocatted can be assessed.
a
Fiigure 3.1 below show thhe
p
project
plannning for roaad constructtion in ordeer to shortenn the trip frrom Chaah to
B
Bukit
Kepon
ng by a distaance of 24 km
m.
Figure 3.1
1: The Projeect Planning for the Roadd Constructions that shorten the trip
between C
Chaah and Bu
ukit Kepongg
45
3.3
Data Collection
The assessment of data collection will be made through two sources as
mention earlier. The MODIS satellite data can be obtained for free by downloading
at the website; http://ladsweb.nascom.nasa.gov. Though, the satellite data can help
in producing CO2 concentration at study areas but in situ data also needed to ensure
the quality of data collection. The special equipment such as Carbon Dioxide (CO2)
Meter for gaseous concentration will be utilized in order to obtain in-situ data. Six
points were allocated at study area in measuring CO2 concentration.
3.4
Carbon Dioxide Meter
For in-situ data, this study will take into consideration the volumes of CO2
being released to the atmosphere. Therefore, the measurement data can be obtained
by using a Carbon Dioxide (CO2) meter which automatically detects the CO2 gasses
that release in atmosphere.
This equipment also includes several features such as indoor air quality
monitor with low-drift NDIR CO2 sensor, user friendly, accurate and reliable CO2
meter and consists of two-line display shows air quality parameters simultaneously.
Indeed that the data measure are in good quality, the appropriate selection for the
model CO2 meter is require.
46
For that reason, this study is carried out using TSI’s Model 7515 IAQ-CALC
which is a cost-effective carbon dioxide (CO2) meter for investigating and
monitoring building air quality and checking ventilation.
There are lists of
specifications that come in this model;
i.
CO2 sensor type
: Dual-wavelength NDIR (non-dispersive infrared)
ii.
Range
: 0 to 5000 ppm
iii.
Accuracy
: ±3.0% of reading or ±50 ppm, whichever greater
iv.
Resolution
: 1 ppm
v.
Response time
: 20 seconds
Figure 3.2: Carbon Dioxide Meter Model 7515.
47
3.5
Algorithm
This study is conducted using level 1B of Terra MODIS satellite data in
Peninsular Malaysia. This data can be obtained from http://ladsweb.nascom.nasa.gov
where three years of MODIS data will be use: 2001, 2005 and 2009. According to a
previous research by Wahid (2008), each of the MODIS data has been geocoded to
the Malaysian Rectified Skew Orthomorpic (MRSO) projection coordinate system.
Based on micrometeorological approach, this model was developed by
Rahman et al., (2004) in order to determine the local scale NPP from MODIS data.
The researchers points out that their new model is capable to track the changing
photosynthetic light use efficiency (LUE) and stress-induced reduction in NPP of
terrestrial vegetation. Based on spectral data, the model is known as simple
“continuous field model” where it only utilizes the visible and near infrared bands
(band 1: 620-670 nm and band 2: 841-876) as well as the “ocean” bands (band 11:
526-536 nm and band 12: 546-556 nm) to explain the variable of flux tower based
daily NPP. Besides that, the most noteworthy is a spectral index called the
Photochemical Reflectance Index (PRI) that is determined and analyzed
approximately using tower-based LUE values. PRI can be defined by equations
below.
PRI = ( 531 - ref) / ( 531+ ref)................................................................. (3.1)
Where, refer to reflectance at the wavelength (nm) expressed by the numeral
subscripts and ref represents a reference wavelength, typically 550 or 570 nm. Thus,
the calculation of MODIS-derived PRI or MODPRI is given as;
MODPRI = ( b11 – b12) / ( b11+ 12)...................................................... (3.2)
48
Where b11 and b12 refer to band 11 and band 12 of MODIS data respectively.
Others than that, by using band 1 and band 2, calculation of the Normalized
Difference Vegetation Index (NDVI) can be made as shown in equation below;
NDVI = ( b2 – b1) / ( b2 + b1).............................................................. (3.3)
After that, the utilization of NDVI values can be apply to calculate the fraction of
PAR (fPAR) absorbed by vegetation using equation is given as;
fPAR = 1.24 x NDVI – 0.168.................................................................... (3.4)
Therefore, by using the relationship of APAR = fPAR x PAR calculation for the
absorbed PAR (APAR) can be determine for each pixel.
The PAR values is
restricted to portion of electromagnetic spectrum from 0.4 to 0.7 μm which is
comparable to the range of light the human eye can see.
Continuously, the
explanation of 88% of the variability can be finding by using linear relationship
between NPP and the (MODPRI x APAR) as shown in following equation;
NPP (C m2/y) = 0.5139 (MODPRI x APAR) – 1.9818............................. (3.5)
3.6
Spectral Bands
The application of MODIS satellite data collection is based on the choice of
spectral bands. Spectral bands can be known when the number of atoms is large, one
gets continuum of energy levels. Besides that, spectral bands also are part of optical
spectra of polyatomic systems, including condensed materials, large molecules etc.
Since this study only involved with terrestrial vegetation, the bands were utilized
were band 1, 2, 11 and band 12 where the primary use for ocean colour, land, cloud
boundaries, phytoplankton and biogeochemistry.
49
Based on previous research by Salomonson and Lawrence (1992), the high
spatial resolution bands (659 nm and 865 nm, centre wavelength) will improve land
boundary and feature detection and will provide enhance cloud versus land
discrimination. However, in the past years these bands have been applied for many
purposes such as in a spectral vegetative index, to study land surface properties with
links to leaf area index, percent plant canopy cover, photosynthetic capacity and
temporally integrated to net primary productivity (NPP), within pixel spectral
mixture modelling and minimum plant canopy resistance to water vapour flux.
Figure 3.3: The Lists of Spectral Bands to Produce MODIS Satellite Image.
50
3.7
ERDAS Software
In this study, the data obtained from MODIS satellite will be analyzed using
ERDAS software. ERDAS is the raster-centric software GIS professionals use to
extract information from satellite and aerial images. Since the MODIS satellite data
perform in imagery data, this software will help to produce the amount of carbon
sequestration by palm oil plantation. Besides that, in order to determine the carbon
value in the proposed location, several algorithms are needed as well as shown in
equation 3.5. Although there are a lot of other software that can be utilized, this
software is chosen because of it was designed specifically for image processing,
practical to apply and comprehensive collection of tools that help to create accurate
base imagery. The study processes is shown in Figure 3.4.
Proposed Location
Measure CO2 concentration
using CO2 Meter Detector
Compare NPP value for
validation process between Insitu and MODIS satellite data
i.
ii.
Data Collection
In-situ
MODIS Satellite
Result and Analysis
Obtain NPP value by using
ERDAS Software
MODIS Satellite Data
Download data from the website;
http://ladsweb.nascom.gov
Figure 3.4: Work Schedule
CHAPTER 4
RESULTS AND DISCUSSION
4.1
Introduction
According to the implementation method that mentioned in Chapter 3, the data
that used is level 1B of MODIS satellite data whereby this data is obtained by
downloading at http://ladsweb.nascom.nasa.gov. This study however is conducted using
three years of MODIS satellite data; 2001, 2005, and 2009, respectively with minimum
cloud cover selected using quick-look menu provided in this website. Furthermore, the
information for these data will be displayed in satellite image as shown in Figure 4.1,
Figure 4.2 and Figure 4.3 below.
52
Figure 4.1: MODIS Satellite Image for Year 2001
53
Figure 4.2: MODIS Satellite Image for Year 2005
54
Figure 4.3: MODIS Satellite Image for Year 2009
55
4.2
The Amount of Net Primary Production and Normalized Difference
Vegetation Index
In order to assess NPP and NDVI, several stages of processing procedures are
implemented using ERDAS Imagine (version 9.1), ENVI (version 4.2) and ArcGIS
(version 9.3). The assessment will be concentrated on a few parameters such as
minimum value, maximum value, mean value and standard deviation value. Therefore,
these parameters hopefully can show the range of changes of carbon values at Felda
Maokil at certain period of years. Based on current condition, since there a road
construction at proposed location, the values for both NPP and NDVI should be lower
than the years before. Table and pie chart below show the values of NPP and NDVI
according to the parameters that required.
Table 4.1: Net Primary Production Values at Felda Maokil in unit of g C m2 /d
YEAR
MIN
MAX
MEAN
STD DEVIATION
2001
2005
2009
222.89
503.55
262.77
569.26
606.05
642.07
468.169
560.685
541.781
88.038
16.770
91.508
Table 4.2: Normalized Difference Vegetation Index Values at Felda Maokil in unitless
YEAR
MIN
MAX
MEAN
STD DEVIATION
2001
2005
2009
0.385
0.738
0.389
0.853
0.892
0.866
0.696
0.863
0.716
0.121
0.029
0.121
56
Stan
ndardDeviation
nforNPP
PValue
2001
2005
2009
45%
47%
8%
Figure 4.4: Standard Deviation
D
forr NPP Valuee at Felda Maaokil (2001, 2005 and 20009)
Stan
ndardDe
eviationforNDV
VIValue
2001
2005
45%
2009
4
44%
11%
D
for NDVI valuee at Felda Maokil
M
(2001, 2005 and 2009)
2
Figure 4.5: Standard Deviation
As shown in Taable 4.1, the NPP value for year 20001 is much lower than 2005
a 2009. From
and
F
the daata in that tabble, the meaan NPP valuue for year 2001
2
gives loowest
m
mean
value which is 4668.169 follow
wed by 560.685 (2005) and 541.7881 (2009). These
T
d
data
look in
nteresting beecause the mean
m
values in both NP
PP and NDV
VI producedd high
v
values.
Bassed on data in
i Table 4.2, we can seee that the higgh mean vallue of NDVI was
57
0.863 which occurred in year 2005 while both year 2001 and 2009 gave the results of
0.696 and 0.716. Further analysis showed that different expectation on standard
deviation. According to Figure 4.4, the percentage of standard deviation for NPP value
in year 2005 (8%) was much lower than 2009 (47%) and 2001 (45%). Meanwhile in
Figure 4.5, there were some changes whereby the percentage of standard deviation for
NDVI value for year 2009 (45%) is higher than 2001 (44%) and 2005 (11%).
The rate of carbon dioxide (CO2) which was absorbed relied on the numbers of
vegetation. The NPP value showed that 2001 was lower than 2005 and this happened
because of age factor. According to the data recorded, palm oil tree in year 2001 was 17
years old compared to 2005 which 21 years old.
Therefore, the palm oil growth
consumed the amount of carbon that was required for the photosynthesis process.
Nevertheless, the NPP value in year 2005 dropped off in year 2009.
It resulted from the numbers of palm oil trees perished at certain period and road
construction that has reduced vegetation at plantation area. The difference between
values of NPP and NDVI by years (2001, 2005 and 2009) can be observed clearly
according to Figure 4.6, Figure 4.7, Figure 4.8, Figure 4.9, Figure 4.10 and Figure 4.11.
From these figures, the colours appeared represent the lower (red) and the higher (blue)
values for both NPP and NDVI. Therefore, these colours can detect the rate of carbon
sink be it higher or lower at the study area.
58
Figure 4.6: The Different Colour of Net Primary Productivity (2001)
59
Figure 4.7: The Different Colour of Net Primary Productivity (2005)
60
Figure 4.8: The Different Colour of Net Primary Productivity (2009)
61
Figure 4.9: The Different Colour of Normalized Difference Vegetation Index (2001)
62
Figure 4.10: The Different Colour of Normalized Difference Vegetation Index (2005)
63
Figure 4.11: The Different Colour of Normalized Difference Vegetation Index (2009)
64
4.3
Correlation between Net Primary Production and Normalized Difference
Vegetation Index Values (2001, 2005 and 2009)
In order to measure air quality at study area, the analysis progressions are
required to correlate the values of NPP and NDVI. The intention of this analysis is to
determine the impact of NDVI and NPP values. Therefore, the significant impacts of
amount carbon that is absorbed by vegetation can be distinguished by ordering NPP and
NDVI values in scattering following different year of 2001, 2005 and 2009.
NDVI(unieless)
NPPversusNDVI(2001)
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0
100
200
300
400
R²=0.990
500
600
NPP(gCm2y1)
Figure 4.12: Scatter Plots of NPP and NDVI values in year 2001 at Felda Maokil
65
NDVI
NPPversusNDVI
R²= 0.120
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
520
540
560
580
600
620
NPP
Figure 4.13: Scatter Plots of NPP and NDVI values in year 2005 at Felda Maokil
NDVI
NPPversusNDVI
R²=0.924
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0
100
200
300
400
500
600
700
NPP
Figure 4.14: Scatter Plots of NPP and NDVI values in year 2009 at Felda Maokil
66
The capability of carbon being absorbed from atmosphere depended on the
numbers of vegetation at the study area. The correspondence between NPP and NDVI
can be detected by observing whether the numbers of vegetation at proposed location is
high or not. The quantity of carbon in the atmosphere will decrease as a consequence of
photosynthesis process by vegetation. From the graph above, we can see that the
relation between NDVI and NPP gave the strongest correlation in year 2001 by referring
to value of R2 = 0.990 as showing in Figure 4.12. Meanwhile, Figure 4.13 shows the
lowest correlation between NPP and NDVI which gives the value of R2 = - 0.120. The
error of correlation value between NPP and NDVI occurred possibly during processing
stage or satellite image itself.
Besides that, there were other elements which influenced the correlation between
NPP and NDVI. The negative correlation rose when there was unstable growth of palm
oil tree. The use of unsuitable fertilizer gave them less nutrient which they require. For
that reason, only a few of palm oil trees attempted to absorb more CO2 as well as the
range of their age could perform.
4.4
Relationship between of APAR and MODPRI at Study Area
The NPP value can be acquired by following the Equation 3.5 in Chapter 3. In
order to evaluate the precision of NPP, the value of MODPRI and APAR were required
to compare their relationship by using Figure 4.17, Figure 4.18 and Figure 4.19. In
practical, if the R-squared value (R2) between MODPRI and APAR is high, so therefore
the quality of NPP value produced was in good condition. Table 4.3 and Table 4.4
below display the results of APAR and MODPRI which attained during analysis stage.
67
T
Table
4.3: APAR
A
values at Felda Maokil
M
YEAR
Y
MIN
MAX
X
ME
EAN
STD
D DEVIATION
2001
2
2005
2
2009
2
904.34
2236.5
1059
2602..6
2810..6
3044..3
20355.088
26999.932
24177.864
438.880
107.117
505.475
Table 4.4: MO
ODPRI valuues at Felda Maokil
M
YEAR
Y
MIN
MAX
X
ME
EAN
STD
D DEVIATION
2001
2
2005
2
2009
2
0.3913
0.3500
0.3809
0.50000
0.48448
0.50000
0.4453
0.4406
0.4443
0.019
0.024
0.027
APA
ARVALUEATDIFFFERENTTYEARS
3500
3000
2500
2000
1500
1000
500
0
2810.6
2602.6
2035.0
088
3044.3
2699.932
2417.864
2236.5
1
1059
904.3
34
2001
2005
MIN
MAX
2009
MEAN
Figure 4.15:
4
A Grapph of APAR
R values at diifferent yearrs
68
MODPRIVALUEATD
DIFFEREN
NTYEARSS
0.5
0.5
0.3913
3
0.5
0.484
48
0.453
0.35
0.4
0.406
0.38
809
0.443
0.3
0.2
0.1
0
2001
2005
MIN
MAX
2009
MEAN
Figure 4..16: A Graphh of MODPR
RI values at different yeaars
As illlustrated in Figure 4.155 above, therre was not much
m
variatioon between mean
v
values
in year 2005 and 2009. On the
t other hannd, the meann value proviided in year 2001
w the low
was
west value coompared to other years.. Furthermoore, Figure 4.16
4
showedd that
t
there
was not as much difference for
f mean vaalue in year 2001, 20055 and 2009. The
v
value
that prresented in year
y 2005 is lower than 2001
2
and 20009.
69
APARvsMODPRI
3000
R²=0.135
2500
APAR
2000
1500
1000
500
0
0
0.2
0.4
0.6
0.8
1
1.2
FPAR
Figure 4.17: Relationship between APAR and MODPRI at Felda Maokil (2001)
APARvsMODPRI
3500
R²=0.878
3000
APAR
2500
2000
1500
1000
500
0
0
0.1
0.2
0.3
0.4
0.5
0.6
MODPRI
Figure 4.18: Relationship between APAR and MODPRI at Felda Maokil (2005)
70
APARvsMODPRI
3500
R²=0.742
3000
APAR
2500
2000
1500
1000
500
0
0
0.1
0.2
0.3
0.4
0.5
0.6
MODPRI
Figure 4.19: Relationship between APAR and MODPRI at Felda Maokil (2009)
The graph linear above displays the relationship between APAR and MODPRI at
Felda Maokil. Based on these graphs, the correlation between APAR and MODPRI can
be explained by referring to R-squared value (R2). The highest R2 value was given in
year 2005 (0.878) and the lowest R2 was given in year 2001 (0.135). Meanwhile for
year 2009, R2 was given as 0.742. The lower value in 2001 was cause by the large
amount of rainfall which is 301.2 mm. For the time being, 2005 and 2009 are not much
difference in R2 value since rainfall amounts of both years being similar with 40.4 mm
(2005) and 49.8 mm (2009). The used of spectral bands in analyze NPP value had been
disturbed by rainfall. It means that, there is might be some noise that contained in
MODIS data for year 2001.
71
4.5
Relationship between Field Measurement Data and Meteorological Data
The assessment of amount of carbon sequester by palm oil trees can be estimated
in several methods. The meteorological data as well as MODIS satellite data can be
verified by compared with field measurement data. The methods which applied are
Reliability test, Student’s t-test, Analysis of Variance (ANOVA) and Pearson correlation
coefficient.
4.5.1
Reliability Test
Stephen (2004) has mentioned that the reliability test refer to the accuracy and
completeness of computer-processed data, given the intended purposes for use.
However, reliability does not mean that computer-processed data is error-free. By
conducting the reliability test, any errors found were within a tolerable range where the
associated risk and found the errors are not significant enough to cause a reasonable
person, aware of the error, to doubt a finding or recommendation based on the data. In
this calculation process, the results for new tolerable ranges are obtained through normal
distribution. Essentially, the total area under the normal curve is 1.00. Yet, further
specific portions of a normal curve lie between plus and minus any given number of
standard deviations from the mean. Therefore, the calculation is applied based on figure
below whereby 99.7% of all values in a normally distributed population lie within ± 3
standard deviation from the mean. But only 0.15% of area under the curve in either side
of the mean value lies outside this range.
72
Figure 4.20: The Range of Area Covered in A Normal Distribution.
Table 4.5: Net Primary Production values at tolerable range
YEAR
MIN
MAX
MEAN
STD DEVIATION
2001
2005
2009
204.055
510.375
267.257
732.283
610.995
816.305
468.169
560.685
541.781
88.038
16.770
91.508
Table 4.6: Normalized Difference Vegetation Index values at tolerable range
YEAR
MIN
MAX
MEAN
STD DEVIATION
2001
2005
2009
0.333
0.776
0.353
1.059
0.950
1.079
0.696
0.863
0.716
0.121
0.029
0.121
73
4.5.2
Student’s t-test
Student’s t-test is operating by comparing the actual difference between two
means in relation to the variation in the data.
Research hypothesis for Student’s t-test;
Ho = There is no difference means between field measurement and meteorological data.
Ha = There is a difference means between field measurement and meteorological data.
The resulted was showed that the tobserved (0.7645) was less than tcritical (2.447)
and it proves the null hypothesis was accepted. This would be concluded there was no
difference between field measurement and meteorological data.
4.5.3 Analysis of Variance (ANOVA)
ANOVA test is conducted in order to test for significance between means.
Moreover, this test also would compare the variability that is being observed between
the two conditions to the variability observed within each condition.
74
i.
NPP
Table 4.7: ANOVA for NPP
Source of
df
Sum of Square,
Mean Squares,
F
Fcritical
Among Group, C
2
5996.46
2998.23
0.337
3.554
Within Group, e
18
160011
8889.51
Total
20
166008
Note: Ho = There is no change in mean value of NPP
Ha = There is a change in mean value of NPP
Based on the table above, observed F value is lower than Fcritical which is 0.337<
3.554. Hence, null hypothesis is accepted which brings the strong evidence to prove and
support that the mean value for each year are the same.
ii.
NDVI
Table 4.8: ANOVA for NDVI
Source of
df
Sum of Square,
Mean Squares,
F
Fcritical
Among Group, C
2
0.0876
0.0438
5.867
3.554
Within Group, e
18
0.1344
0.00747
Total
20
0.2220
Note: Ho = There is no change in mean value of NPP
Ha = There is a change in mean value of NPP
Based on the table above, observed F value is higher than Fcritical which is 5.867 >
3.554. Hence, null hypothesis is rejected which brings conclusion that the mean value
for each year is not the same.
75
4.5.4
Pearson’s Correlation Coefficient
Pearson’s Correlation Coefficient is also known as Karl Pearson’s Correlation.
This method was used in measured the correlation between two variables X and Y
whereby the range value is between +1 and -1 inclusively.
A value of +1 indicated that there was a linear relationship between X and Y
perfectly with all data points lied on a line. Meanwhile, a value of -1 indicated that all
data points lied on a line for which Y decrease as X increases. However, a value of 0
indicates that there is no linear relationship between variables.
Table 4.9: Correlation between NPP and NDVI
Year
2001
2005
2009
Pearson Correlation
0.581
-0.532
0.962
Probable Error
0.019
0.021
0.002
Based on Table 4.9, only 2005 were shown negative correlation in its data. This
test proved that the Figure 4.13 had negative correlation between NPP and NDVI.
However, 2001 and 2009 consists of data which had positive correlation. The different
data obtained between 2001, 2005 and 2009 are cause by three factors such as the
amount of rainfall, age of oil palm trees and types of fertilizers.
The age of oil palm trees was 17 in 2001 and this give moderate grow of oil palm
trees with the rainfall fall amounts was 301.2 mm and the used of fertilizers are suitable
with the conditions of trees. When come to 2005, the age of oil palms was 21 but the
utilization of available fertilizers was not accepted.
76
Therefore, there are unwell grow among them with decreased of rainfall amounts
40.4 mm. However, only a few of oil palm trees that give a good performance in
sequester carbon dioxide which makes the result negative correlation. In 2009, the used
of fertilizers had changed since 2008. The age of oil palm trees was 25 which required
some changing in characteristic of their fertilizers. Because of that reason and enough
water with 49.8 mm of rainfall amounts, most of oil palm trees are enables to sequester
carbon dioxide. This situation has leaded to the highest correlation between NPP and
NDVI.
4.6
Effects of Road Construction in Carbon Absorption
The analysis of MODIS data were presented some changed to the amount of Net
Primary Productivity (NPP) affected from road construction. Based on the figures
below, the quantities of carbon sequester decreased and these can be detected by
following the transformation of colours at proposed area. The images displayed were
the group of Felda Maokil which included Felda Maokil 1, Felda Maokil 2, Felda
Maokil 3 and Felda Maokil 4. However, the studied area was located at Felda Maokil 1
and Felda Maokil 2 related to the road development. Therefore, the studied area can be
recognized by the line on images which indicated as proposed road.
The colours on images represented the quantity of carbon dioxide (CO2) at oil
palm trees and vegetation index at oil palm plantation. The blue colour corresponds to
highest NPP value whereas the red colour responds to lowest NPP value. Therefore, the
colours obtained were red in colour indicating that both values of NPP and NDVI were
low in year 2009.
77
By comparing both images, it had proved that there were relationships between
NPP and NDVI. This is because, the colours turned red only at position where there was
a road which under construction.
The understanding of implication from road
development where the oil palm trees which acted as carbon sequester had been cut
down. As a result, NPP which is known as carbon absorbed by plants were low as a
result of numbers of oil palm tree being reduced.
78
Figure 4.21: Location of Road Construction at Study Area (2001) based on NPP
79
Figure 4.22: Location of Road Construction at Study Area (2005) based on NPP
80
Figure 4.23: Location of Road Construction at Study Area (2009) based on NPP
81
Figure 4.24: Location of Road Construction at Study Area (2001) based on NDVI
82
Figure 4.25: Location of Road Construction at Study Area (2005) based on NDVI
83
Figure 4.26: Location of Road Construction at Study Area (2009) based on NDVI
CHAPTER 5
CONCLUSION AND RECOMMENDATION
5.1
Conclusion
This study was conducted to estimate the Net Primary Productivity (NPP) at
Felda Maokil for three years using MODIS images. Four years interval had been
selected in order to figure more variation between each year starting with 2001, 2005
and 2009. The analysis were carried out using model that developed by Rahman et al.
(2004). The model applied is capable to track the changing photosynthetic light use
efficiency (LUC) and stress-induced reduction in NPP of terrestrial vegetation.
85
There are two sources that have been implemented for gather the NPP value
which are Carbon Dioxide (CO2) Meter and MODIS satellite image. However, the NPP
value is unable to get for on-side measurement for year 2001 and 2005 since there are no
recorded data made. Therefore, only current year was achieved to acquire by positioned
seven sampling points. The used of Student’s t-test was showed that the NPP value
from both methods were classified as the same.
On the other hand, three years of MODIS satellite images were running on
ANOVA test. The test was perform in purpose for verify the mean NPP value was the
same. Thus, we can see the pattern appeared based on the value of NPP. The result had
stated that three years MODIS satellite images were produced the same NPP value.
Because of that reason, we can conclude the amount of NPP does not depend on the
numbers of vegetation. If the vegetation grows well, therefore the NPP value is high. A
few factors have been identified which lead to the unexpected result such as rainfall
amount, age of tree and the types of fertilizer used. As a conclusion, the study results
meet the set objectives.
86
5.2
Recommendation
This study however can be improved to set better results in the future. These are
some recommendations suggested in order to make this field of study more successful.
1) The area of study should be larger in order to make sure the data from MODIS
satellite image will be specific or by using satellite with higher spatial resolution
that will help for smaller study areas. Besides that, some comparison is required
between vegetation area and industrial area. By doing that, the effect from
carbon sequestration can be seen clearly.
2) The analysis for MODIS image is done for each year since 2001 to 2009. Thus,
the pattern of NPP can be identified based on the differences for each year.
3) The numbers of sampling points need to be increased in term of variance and
accuracy of the sampling data. Hence, R2 (r-squared) value yields better results.
4) The available NPP data can be utilized for future prediction in order to
investigate the impact of road development to carbon sequestration for long term
period.
According to the points of recommendations above, the results produced should be
much better than available data. Therefore some information needs to be taken under
consideration such as the area of study area, satellite resolution and points of data
measurement.
87
REFFERENCES
1) Barrie, L.A. and R.S. Schemenauer (1986). Pollutant wet deposition mechanisms
in precipitation and fog water. Water Air Soil Pollu., 30, 90-104.
2) Chris Marnay, Diane Fisher, Scott Murtishaw, Amol Phadke, Lynn Price, Jayant
Sathaye (2002). Estimating Carbon Dioxide emissions Factors for the California
Electric Power Sector. Lawrence Berkeley National Laboratory.
3) Deevey, E. S., Jr., (1973). Sulfur, nitrogen and carbin in atmosphere. Carbon and
the Biosphere. In. G. M. Woodwell and E. V. Peacan, (Eds.) U. S. Atomic Energy
Commission, CONF-720510 (pp. 182-190). D.C: Washington.
4) Delwiche, C. C., and Likens, G. E., (1977). Biological response to fossil fuel
combustion products. In W. Stumm. (Ed.) Global Chemical Cycles and Their
Alternations by Man (pp. 73-88). Berlin: Springer-Verlag.
5) Etherington, J.R. (1975) Environment and Plant Ecology. London: John Wiley
and Sons.
6) Ellenberg, H. and Mueller-Dombois, D. Tentative physiognomic-ecological
classification of plant formations of the earth. Ber. Geobot. Inst. Eth. Stiftg.
Rubel, 1967, 37: 21-55.
7) Gifford, R. M. and Musgrave, R. B. Diffusion and quasidiffusion resistances in
relation to the carboxylation kinetics of maize leaves. Physiol. Plant, 1970, 23:
1048-1056.
8) Gourdian. J. and Ajtay, G. L. (1979) The Possible Effects of Increased CO2 on
Photosynthesis. Scientific Committee on Problems of the Environment (SCOPE).
Germany: Ratzeburg.
9) Goudriaan, J. and van Laar, H. H. Measurements of some relations between leaf
resistance, CO2 concentration and CO2 assimilation in maize, beans, lalanggrass
and sunflower. Photosynthetica, 1978, 12: 241-249.
88
10) Hambridge, G. (1949). Hunger Signs in Crops. U.S.A: The National Fertilizer
Association.
11) Kueh Hsiao Chin (2005). Chlorophyll-A Mapping From The Moderate Resolution
Imaging Spectrometer Data In The South China Sea. Universiti Teknologi
Malaysia.
12) Loomis, R. S. and Gerakis, P. A. (1975) Productivity of agricultural ecosystems.
In Cooper, J. P. (Ed) Photosynthesis and Productivity in Different Environments.
IBP 3. London: Cambridge Univ. Press.
13) Lieth, H. and Whittaker, R. H. (eds) (1975). Primary Productivity of the
Biosphere. Ecol. Stud. 14, 1-339. Springer-Verlag, Berlin, Heidelberg, New York.
14) Lof, H. (1976). Water use efficiency and competition between arid zone annuals,
especially the grasses. Phalaris minor and Hordeum murinum. Agric. Res.
Reports. Wagenengen: Pudoc, 853.
15) Lindberg, S.E. and R.C. Harriss, and R.R. Turner. Atmospheric deposition of
metals to forest vegetation. Science, 1982, 215: 1609-1011.
16) Likens, G. E., Bormann, H. F., and Hohnson, N. M. (1981). Interactions between
major biogeochemical cycles in terrestrial ecosystems. In G. E. Likens, (Ed) Some
Perspectives of the Major Biogeochemical Cycles. (pp. 93-112). New York:
SCOPE 17, Wiley.
17) Mohd Syaiful Bin Sutan Shahril (2008). Estimation of Chrorophyll-A
Concentration and Monitoring Chlorophyll Pattern From Satellite MODIS Data.
Universiti Teknologi Malaysia
18) Natr, L. (1975) Influence of mineral nutrition on photosynthesis and the use of
assimilates In Cooper, J. P. (Ed) Photosynthesis and Productivity in Different
Environments. London: IBP 3, Cambridge Univ. Press.
19) Ovington, J.D. (1965) Woodlands. Modern Biology Series, 1-574. The English
Univ. Press, London.
20) Pearce, J. M., Holloway, S., Wacker, H., Nelis, M. K., Rochelle, C. A. &
Bateman, K. (1996). Natural occurrences as analogues for carbon dioxide
disposal. Energy Convers. Manage. 37: 1123-1128.
21) Price, J., Smith B. (2008). Geologic Storage of Carbon Dioxide-Staying Safely
Underground. IEA Greenhouse Gas R&D Programme.
89
22) Redfield, A. C., Ketchum, B. H., and Ridchard, F. A., 1963, The influnce of
organisms on the composition of seawater. In M. N. Hill., (Ed.) The Sea, v. 2 (pp.
26-77). New York: Wiley.
23) Sam Holloway (2007). Carbon Dioxide Capture and Geologic Storage.
Philosophical Transactions of The Royal Society A. 365, 1095-1107
24) Studlick, J. R. J., Shew, R. D., Basye, G. L. & Ray, J. R. (1990). A giant carbon
dioxide accumulation in the Norphlet Formation, Pisgah Anticline, Mississipi. In
J. Barwis, J. G. McPherson & J. R. J. Studlick., (eds.) Sandstone petroleum
reservoirs (pp. 181-203). New York, NY: Springer.
25) Stevens, S. H. (2005) Natural CO2 fields as analogs foe geologic CO2 storage. In
D. C. Thomas & S. M. Benson (Eds.) Carbon dioxide capture for storage in deep
geologic formations-results from CO2 capture project, vol. 2 (pp. 687-697).
Oxford, UK: Elsevier.
26) Schemenauer, R.S. (1986). Acidic deposition to forests: The 1985 chemistry of
high elevation fog (CHEF) project. Atmos. Ocean 24: 303-328.
27) Trabalka, J.R., J.A. Edmonds, J.M. Reilly, R.H. Gardner, and D.E. Reichle (1986)
Atmospheric CO2 projections with globally averaged carbon cycle models. In
J.R. Trabalka and D.E. Reichle, (Eds.) The Changing Carbon Cycle of Global
Analysis (pp. 534-560). New York: Springer-Verlag.
28) U.S. Environmental Protection Agency (1976). Diagnosing Vegetation Injury
Caused by Air pollution. Contract No. 68-02-1344, U.S.E.P.A., Air Pollution
Training Institute. NC: Reseasch Triangle Park.
29) UNESCO (1973). International Classification and Mapping of Vegetation, 1-93.
Paris: UNESCO.
30) V. V. Salomonson, D. L. Toll and W. T. Lawrence (1992). Moderate Resolution
Imaging Spectrometer (MODIS) and Observations of The Land Surface.
NASA/Goddard Flight Center. USA: Greenbelt.
31) Walter, H. (1973) Vegetationszonen and Klima, 2nd ed., 1-253. Ulmer Verlag,
Stuttgart.
32) Weathers, K.C., G.E. Likens, F.H. Bormann, S.H. Bicknell, B.T. Bormann, B.
Daube Jr., J.S. Eaton, J.N. Galloway, W.C. Keene, K.D. Kimball, W.H.
MacDowell, T.G. Siccaina, D. Smiley, and R.A. Tarrant (1988). Claoud water
chemistry from ten sites in North America. Environ. Sci. Technol.
90
APPENDIX A
Manual Calculation for Reliability Test
NPP
2001
ܵ‫݀ݐ‬Ǥ ‫ ݒ݁ܦ‬ൌ ͺͺǤͲ͵ͺ
ͺͺǤͲ͵ͺ ൈ ͵ ൌ ʹ͸ͶǤͳͳͶ
‫ ‹ ׵‬ൌ ݉݁ܽ݊ െ ʹ͸ͶǤͳͳͶ ൌ Ͷ͸ͺǤͳ͸ͻ െ ʹ͸ͶǤͳͳͶ ൌ ʹͲͶǤͲͷͷ
‫ šƒ ׵‬ൌ ݉݁ܽ݊ ൅ ʹ͸ͶǤͳͳͶ ൌ Ͷ͸ͺǤͳ͸ͻ ൅ ʹ͸ͶǤͳͳͶ ൌ ͹͵ʹǤʹͺ͵
2005
ܵ‫݀ݐ‬Ǥ ‫ ݒ݁ܦ‬ൌ ͳ͸Ǥ͹͹Ͳ
ͳ͸Ǥ͹͹Ͳ ൈ ͵ ൌ ͷͲǤ͵ͳͲ
‫ ‹ ׵‬ൌ ݉݁ܽ݊ െ ͷͲǤ͵ͳͲ ൌ ͷ͸ͲǤ͸ͺͷ െ ͷͲǤ͵ͳͲ ൌ ͷͳͲǤ͵͹ͷ
‫ šƒ ׵‬ൌ ݉݁ܽ݊ ൅ ͷͲǤ͵ͳͲ ൌ ͷ͸ͲǤ͸ͺͷ ൅ ͷͲǤ͵ͳͲ ൌ ͸ͳͲǤͻͻͷ
2009
ܵ‫݀ݐ‬Ǥ ‫ ݒ݁ܦ‬ൌ ͻͳǤͷͲͺ
ͻͳǤͷͲͺ ൈ ͵ ൌ ʹ͹ͶǤͷʹͶ
‫ ‹ ׵‬ൌ ݉݁ܽ݊ െ ʹ͹ͶǤͷʹͶ ൌ ͷͶͳǤ͹ͺͳ െ ʹ͹ͶǤͷʹͶ ൌ ʹ͸͹Ǥʹͷ͹
‫ šƒ ׵‬ൌ ݉݁ܽ݊ ൅ ʹ͹ͶǤͷʹͶ ൌ ͷͶͳǤ͹ͺͳ ൅ ʹ͹ͶǤͷʹͶ ൌ ͺͳ͸Ǥ͵Ͳͷ
91
NDVI
2001
ܵ‫݀ݐ‬Ǥ ‫ ݒ݁ܦ‬ൌ ͲǤͳʹͳ
ͲǤͳʹͳ ൈ ͵ ൌ ͲǤ͵͸͵
‫ ‹ ׵‬ൌ ݉݁ܽ݊ െ ͲǤ͵͸͵ ൌ ͲǤ͸ͻ͸ െ ͲǤ͵͸͵ ൌ ͲǤ͵͵͵
‫ šƒ ׵‬ൌ ݉݁ܽ݊ ൅ ͲǤ͵͸͵ ൌ ͲǤ͸ͻ͸ ൅ ͲǤ͵͸͵ ൌ ͳǤͲͷͻ
2005
ܵ‫݀ݐ‬Ǥ ‫ ݒ݁ܦ‬ൌ ͲǤͲʹͻ
ͲǤͲʹͻ ൈ ͵ ൌ ͲǤͲͺ͹
‫ ‹ ׵‬ൌ ݉݁ܽ݊ െ ͲǤͲͺ͹ ൌ ͲǤͺ͸͵ െ ͲǤͲͺ͹ ൌ ͲǤ͹͹͸
‫ šƒ ׵‬ൌ ݉݁ܽ݊ ൅ ͲǤͲͺ͹ ൌ ͲǤͺ͸͵ ൅ ͲǤͲͺ͹ ൌ ͲǤͻͷͲ
2009
ܵ‫݀ݐ‬Ǥ ‫ ݒ݁ܦ‬ൌ ͲǤͳʹͳ
ͲǤͳʹͳ ൈ ͵ ൌ ͲǤ͵͸͵
‫ ‹ ׵‬ൌ ݉݁ܽ݊ െ ͲǤ͵͸͵ ൌ ͲǤ͹ͳ͸ െ ͲǤ͵͸͵ ൌ ͲǤ͵ͷ͵
‫ šƒ ׵‬ൌ ݉݁ܽ݊ ൅ ͲǤ͵͸͵ ൌ ͲǤ͹ͳ͸ ൅ ͲǤ͵͸͵ ൌ ͳǤͲ͹ͻ
Summary of NPP
YEAR
2001
2005
2009
MIN
204.055
510.375
267.257
MAX
732.283
610.995
816.305
STD.DEVIATION (3)
264.144
50.310
274.524
Summary of NDVI
YEAR
2001
2005
2009
MIN
0.333
0.776
0.353
MAX
1.059
0.950
1.079
STD.DEVIATION (3)
0.363
0.087
0.363
92
APPENDIX B
Manual Calculation for Student’s t-test
Field Measurement
460
468
401
404
395
472
401
Meteorological Data
517.457
401.944
533.906
384.765
364.641
262.770
315.855
Different (࢞)
-57.457
66.056
-132.906
19.235
30.359
209.23
85.145
ܶ‫ ݔ݈ܽݐ݋‬ൌ ʹͳͻǤ͸͸ʹ
ܶ‫ ݔ݈ܽݐ݋‬ଶ ൌ ͹͹͸Ͷ͹Ǥʹʹͷ
σ‫ݔ‬
݊
ʹͳͻǤ͸͸ʹ
ൌ
ൌ ͵ͳǤ͵ͺͲ
͹
‫ݔ‬ҧ ൌ σሺ‫ ݔ‬െ ‫ ݔ‬ଶ ሻଶ
͹͹͸Ͷ͹Ǥʹʹͷ
ൌඨ
ൌ ͳͳ͵Ǥ͹ͷͻ
͸
݊െͳ
ܵ‫݊݋݅ݐܽ݅ݒ݁ܦ݀ݎܽ݀݊ܽݐ‬ǡ ‫ ݏ‬ଶ ൌ ඨ
࢞૛
3301.307
4363.395
17664
369.9852
921.6689
43777.19
7249.671
93
͹͹͸Ͷ͹Ǥʹʹͷ
͸
ൌඨ
ൌ ͳͳ͵Ǥ͹ͷͻ
In order to estimate standard error between two mean, the equation as showing below;
ܵመ ൌ
ൌ
ටσ ൬
‫ݔ‬ҧ ଶ
‫ݔ‬ଶ
൰െ݊൬
൰
݊െͳ
݊െͳ
ξ݊
ሾඥሺ͹͹͸Ͷ͹ǤʹʹͷȀ͸ሻ െ ͹ሺ͵ͳǤ͵ͺͲଶ Ȁ͸ሻሿ
ξ͹
Observed ‫ ݐ‬ൌ ሺ‫ݔ‬ҧ െ ‫݋ܪݔ‬ሻȀ‫ݏ‬Ƹ
ൌ ͵ͳǤ͵ͺͲȀͶͳǤͲͶͶ
ൌ ͲǤ͹͸Ͷͷ
ൌ ͶͳǤͲͶͶ
94
APPENDIX C
Manual Calculation for Analysis of Variance (ANOVA)
i.
NPP
In this case we wish to test the null hypothesis that the population means from these
three sets are equal.
H0 = µ1 = µ2 = µ3
Ha = Not all µ are equal
Location / Year
Point 1
Point 2
Point 3
Point 4
Point 5
Point 6
Point 7
Total
Years
2001
2005
2009
2001
645.7678
639.622
823.6254
407.2125
448.9191
538.124
672.1799
4175.451
2005
530.1532
556.5192
576.1801
552.3281
554.4752
553.7714
581.3975
3904.825
Numbers of observations, ࢔࢐
7
7
7
2009
454.2015
451.3459
595.21
618.9161
620.2216
618.8461
591.7644
3950.506
ഥ࢐
Mean, ࢞
596.493
557.832
564.357
95
N=21
‫ݔ‬ҧ ൌ
σ ‫ݔ‬ҧ௝
ൌ ͷ͹ʹǤͺͻͶ
݊
Where;
c = the number of group
‫ݔ‬௜ೕ = the ith observation in the jth group
݊௝ =the number of observation in group j
N = total number of observations in all group combined
‫ݔ‬ҧ = overall mean
‫ݔ‬ҧ௝ =group mean
‫ݔ‬௜௝ = individual value
ഥሻ૛
ഥ࢐ െ ࢞
SSC = σ࡯ࡶୀ૚ ሺ࢞
ൌ ͹ሾሺͷͻ͸ǤͶͻ͵ െ ͷ͹ʹǤͺͻͶሻଶ ൅ ሺͷͷ͹Ǥͺ͵ʹ െ ͷ͹ʹǤͺͻͶሻଶ ൅ ሺͷ͸ͶǤ͵ͷ͹ െ ͷ͹ʹǤͺͻͶሻଶ ሿ
ൌ ͹ሾͷͷ͸Ǥͺͻͷ ൅ ʹʹ͸Ǥͺ͹ͳ ൅ ͹ʹǤͺ͹ሿ ൌ ͷͻͻ͸ǤͶ͸
࢔
࢏
σ࡯ࡶୀ૚ ሺ࢞࢏࢐ െ ࢞
ഥ ࢐ ሻ૛
SSE = σ࢏ୀ૚
ൌ ʹͶʹͺǤͲͳଶ ൅ ͳͺ͸ͲǤͳͳଶ ൅ ͷͳͷͺͻǤʹଶ ൅ ͵ͷͺʹ͹Ǥͳଶ ൅ ʹͳ͹͹ͺଶ ൅ ͵ͶͲ͸Ǥͻ͵ଶ ൌ ͳ͸ͲͲͳͳ
൅ͷ͹ʹͺǤͷͳଶ ൅ ͹͸͸Ǥͳʹʹଶ ൅ ͳǤ͹ʹͶଶ ൅ ͵͵͸Ǥ͸Ͷ͹ଶ ൅ ͵ͲǤʹͻͶଶ ൅ ͳͳǤʹ͸ͻଶ ൅ ͳ͸ǤͶͻͲଶ
൅ͷͷͷǤ͵ʹͻଶ ൅ ͳʹͳ͵ͶǤͶଶ ൅ ͳʹ͹͹ͳǤ͹ଶ ൅ ͻͷͳǤͺͷଶ ൅ ʹͻ͹͸Ǥ͸ଶ ൅ ͵ͳʹͲǤ͹ͷଶ
൅ʹͻ͸ͺǤͻ͸ଶ ൅ ͹ͷͳǤͳͳͶଶ
SST = SSC + SSE = 166008
Degree of Freedom
Degree of Freedom , df
dfc
c-1
2
dfe
n-c
18
dft
n-1
20
96
Mean Squares of Among Group =
Among Groups Sum of Squares SSC
=
Among Group degree of freedom dfC
Mean Squares of Within Group =
Within Group of Squares
SSE
=
Within Group degree of freedom dfe
Source of
Among Group, C
Within Group, e
Total
ii.
df
2
18
20
Sum of Square,
5996.46
160011
166008
Mean Squares,
2998.23
8889.51
F
0.337
Fcritical
3.554
NDVI
In this case we wish to test the null hypothesis that the population means from these
three sets are equal.
H0 = µ1 = µ2 = µ3
Ha = Not all µ are equal
Location / Year
Point 1
Point 2
Point 3
Point 4
Point 5
Point 6
Point 7
Total
2001
0.84
0.8064
0.7984
0.5741
0.6282
0.7226
0.7741
5.1441
2005
0.892
0.888
0.824
0.882
0.885
0.884
0.848
6.104
2009
0.589
0.584
0.746
0.831
0.758
0.860
0.775
5.147
97
Numbers of observations, ࢔࢐
7
7
7
Years
2001
2005
2009
‫ݔ‬ҧ ൌ
N=21
ഥ࢐
Mean, ࢞
0.734
0.872
0.735
σ ‫ݔ‬ҧ௝
ൌ ͲǤ͹ͺͳ
݊
ഥሻ૛
ഥ࢐ െ ࢞
SSC = σ࡯ࡶୀ૚ ሺ࢞
ൌ ͹ሾሺͲǤ͹͵ͷ െ ͲǤ͹ͺͳሻଶ ൅ ሺͲǤͺ͹ʹ െ ͲǤ͹ͺͳሻଶ ൅ ሺͲǤ͹͵ͷ െ ͲǤ͹ͺͳሻଶ ሿ
ൌ ͹ሾͲǤͲͲʹͳͳ ൅ ͲǤͲͲͺ͵Ͷ ൅ ͲǤͲͲʹͲ͸ሿ ൌ ͲǤͲͺ͹͸
࢔
࢏
σ࡯ࡶୀ૚ ሺ࢞࢏࢐ െ ࢞
ഥ ࢐ ሻ૛
SSE = σ࢏ୀ૚
ൌ ͲǤͲͳͳͲͷଶ ൅ ͲǤͲͲͷͳʹଶ ൅ ͲǤͲͲͶͲͶଶ ൅ ͲǤͲʹͷͺʹଶ ൅ ͲǤͲͳͳ͵ͺଶ ൅ ͲǤͲͲͲͷͷଶ
൅ͲǤͲͲͳͷͷଶ ൅ ͲǤͲͲͲͶͳଶ ൅ ͲǤͲͲͲʹͷଶ ൅ ͲǤͲͲʹ͵ʹଶ ൅ ͲǤͲͲͲͳଶ ൅ ͲǤͲͲͲͳ͹ଶ
൅ͲǤͲͲͲͳͷଶ ൅ ͲǤͲͲͲͷͷଶ ൅ ͲǤͲʹͳ͵ͳଶ ൅ ͲǤͲʹʹ͸Ͷଶ ൅ ͲǤͲͲͲͳͳଶ ൅ ͲǤͲͲͲͻ͵ʹଶ
൅ͲǤͲͲͲͷͶଶ ൅ ͲǤͲͳͷ͹ͷଶ ൅ ͲǤͲͲͳ͸Ͷଶ
ൌ ͲǤͳ͵ͶͶ
SST = SSC + SSE = 0.2220
Source of
Among Group, C
Within Group, e
Total
df
2
18
20
Sum of Square,
0.0876
0.1344
0.2220
Mean Squares,
0.0438
0.00747
F
5.867
Fcritical
3.554