Monitoring land cover changes in tien hai nature reserve thai binh province using multi temporal landsat data

This study is carried out with aim was to monitoring land cover change in Tien Hai Nature Reserve from 2001 to 2015 using multi-temporal Landsat data combine with GIS technique.. Multi-t

ACKNOWLEDGEMENT

This thesis would not have been possible without the support and help from my teachers, friends, families and several people I would like to express our special appreciation of following people who supported me with my sincere gratitude:

I would like to express my sincere thanks and appreciation to my supervisor Dr Nguyen Hai Hoa for his untiring and excellent guidance, valuable suggestions in my dilemmas His comments and advices have helped me to finish my thesis

I am also thankful to Prof Lee MacDonald for his enthusiasm in guiding me to construct thesis proposal His comments and criticism on my proposal helped me to present better final thesis

Further, I would like to thank the Department of Natural Resources and Environment in Tien Hai district for support and giving me the chance to study in Tien Hai Nature Reserve Besides, I also thank the local authorities and famers in three communes (Nam Hung, Nam Phu, Nam Thinh) for providing valuable information and data of the area

I wish to thank the Center of Laboratory and Practice of Vietnam Forestry University for providing me with survey instrument (GPS Garmin) for field survey I also thank my friends for their enthusiastic help in the process collected data in field survey

Last but not least, my everlasting gratitude goes to my parents who always encouraged and supporting me all the time

TABLE OF CONTENTS

ACKNOWLEDGEMENT

TABLE OF CONTENTS

ABBREVIATIONS

LIST OF TABLES

LIST OF FIGURES

ABSTRACT 1

Chapter I INTRODUCTION 3

Chapter II LITERATURE REVIEW 6

2.1 Applications of GIS and Remote Sensing to monitor land use and land cover 6

2.1.1 Land use and land cover 6

2.1.2 GIS - Remote Sensing and monitoring land use - land cover 6

2.1.3 Studies of application of GIS and Remote Sensing to detect land use and land cover 7

2.2 Studies of land use and land cover in Tien Hai Nature Reserve 10

2.2.1 Key study 10

2.2.2 Key significant issues in Tien Hai Nature Reserve 10

Chapter III STUDY GOAL, SPECIFIC OBJECTIVES, METHODOLOGY 12

3.1 Study goal 12

3.2 Specific objectives 12

3.3.1 Data collection 14

3.3.2 Integration of multi-temporal Landsat data and GIS 15

5.2.5 Change detection 23

Chapter IV STUDY SITE, NATURAL AND SOCIO-ECONOMIC FEATURES 24

4.1 Study site 24

4.2 Natural features 24

Chapter V RESULTS AND DISCUSSION 28

5.1 Current status of coastal land cover and its management schemes 28

5.1.1 Current status of coastal land cover in Tien Hai Nature Reserve 28

5.1.2 Management schemes of Tien Hai Nature Reserve 29

5.2 Changes in coastal land covers in Tien Hai Nature Reserve from 2001 to 2015 31

5.2.1 Thematic coastal land covers in Tien Hai Nature Reserve 32

5.2.2 Accuracy assessment of coastal land cover classification 39

5.2.3 Spatial land cover change analysis during 2001 - 2015 40

5.3 Key drivers of land cover change in the Tien Hai Nature Reserve 49

5.3.1 Human - induced drivers and coastal land cover change 49

5.3.2 Natural - induced forces and coastal land cover change 54

5.4 Solutions for better management of coastal land use in Tien Hai Nature Reserve 55

Chapter VI GENERAL CONCLUSION, LIMITATIONS AND FURTHER STUDY 58

6.1 General conclusion 58

6.2 Limitations 59

6.3 Further studies 60 REFERENCES

APPENDICES

ABBREVIATIONS

ASTER Advanced Spaceborne Thermal Emission and Reflection Radiometer ENVI Exelis Visual Information Solutions

ETM+ Enhanced Thematic Mapper Plus

GCPs Ground control points

GIS Geographic Information System

GPS Geographic Position System

IBA Important Bird and Biodiversity Area

IUCN International Union for Conservation of Nature

MARD Ministry of Agriculture and Rural Development

MCD Centre for Marinelife Conservation and Community Development MODIS Moderate-resolution Imaging Spectroradiometer

MSS Multispectral Scanner System

NASA National Aeronautics and Space Administration

NDVI Normalized Difference Vegetation Index

NGO Non-Governmental Organization

OLI Operational Land Imager

SPOT Satellite Pour l’Observation de la Terre

TIF Tagged Image File Format

TIRS Thermal Infrared Sensor

UNESCO United Nations Educational, Scientific and Cultural Organization

UTM Universal Transverse Mercator

LIST OF TABLES

Table 3.1: Satellite images used for the study 17

Table 3.2: Ground control points (GCPs) and land cover classes 18

Table 3.3: Description of land cover classification used 21

Table 5.1: Summary of area of land cover classes from 2001 to 2015 (ha) 32

Table 5.2: Error matrix of accuracy assessment for image classification in 2015 39

Table 5.3: Changes in area of land cover classes in different periods 41

Table 5.4: Change detection of land cover classes from 2001 to 2006 (ha) 42

Table 5.5: Change detection of land cover classes from 2006 to 2015 (ha) 45

Table 5.6: Change detection of land cover from 2001 to 2015 (ha) 47

LIST OF FIGURES

Figure 3.1: Overview flowchart of study objectives 13

Figure 3.2: Flowchart of methodology for image classification and change mapping 16

Figure 3.3: Landsat images and their band combinations used in this study 20

Figure 4.1: Location of study site in Tien Hai district 27

Figure 5.1: Land cover classes for 2001, 2006 and 2015 (ha) 32

Figure 5.2: Area of land cover classes in 2001 33

Figure 5.3: Thematic land cover of Tien Hai Nature Reserve in 2001 34

Figure 5.4: Areas of land cover classes in 2006 (ha) 35

Figure 5.5: Thematic land cover of Tien Hai Nature Reserve in 2006 36

Figure 5.6: Area of land cover classes (ha) in 2015 37

Figure 5.7: Thematic land cover of Tien Hai Nature Reserve in 2015 38

Figure 5.8: Changes in area of land cover classes from 2001 to 2015 (ha) 42

Figure 5.9: Land cover changes in Tien Hai Nature Reserve from 2001 to 2006 44

Figure 5.10: Land cover changes in Tien Hai Nature Reserve from 2006 to 2015 46

Figure 5.11: Land cover changes in Tien Hai Nature Reserve from 2001 to 2015 48

ABSTRACT

Land cover is considered as a fundamental component for sustainable management

of natural resource and monitoring environmental change In recent years, with increasing

in anthropology activities and climate change, land cover is dramatically changed; and impact adversely affected to environment Tien Hai Nature Reserve is coastal wetland zone, has a vital role in contributed immeasurable socio-economic, cultural, and environmental value at local, provincial regional, national level Therefore, Tien Hai Nature Reserve is facing with decline in function and natural resource of the area by human activities and natural calamity This study is carried out with aim was to monitoring land cover change in Tien Hai Nature Reserve from 2001 to 2015 using multi-temporal Landsat data combine with GIS technique From that, would have an overview about land cover change which contributed better management of the Tien Hai Nature Reserve

For this study, multi-temporal Landsat imagery includes Landsat 5 TM, Landsat 7 ETM+ and Landsat 8 OLI/TIRS were analyzed using ArcGIS software; and Global Position System (GPS) was acquired for collected ground control points in the field Iso cluster unsupervised classification was chosen to extract information from satellite data with five land cover classes were identified including water, built-up/sand, casuarinas, mangroves; and others The result of the classification classes given 85.6% accuracy in

2015 The change detection of land cover is divided into 2 periods (2001-2006 and 2015) In overall, casuarinas and built-up/sand slightly changed, and have tended to increased during 14 years Water, mangroves, others have largely changed Mangroves area decreased 245.6 ha in 2001 to 2006, after increased 331.3 ha from 2006 to 2015 Area coverage of water increased from 2001 to 2006, decreased from 2006 to 2015 In contrary, others area decreased from 2001 to 2006, increased from 2006 to 2015

2006-The data on driving factors of land cover were obtained by interviews 2006-The factors

of driver land cover change of Tien Hai Nature Reserve included both human activities and natural calamity However, human induced is mainly contributed with activities is represented by awareness of local people and management plans of board management In addition, natural factors such as storm, sea wave, sea level rise, also drive of change in land cover in Tien Hai Nature Reserve

Chapter I INTRODUCTION

Most current land use and land cover patterns on the earth reflect the interaction of human activities and the natural environment However, alteration of land cover takes place primarily because of anthropological activities rather than natural causes Land cover is understood as surface cover on the ground of land, including water, vegetation, bare soil, and artificial structures (Dewan and Yamahuchi, 2009) Human driven forces such as agricultural, mining, deforestation, construction that influence on shifting patterns of land use and land cover (Dewan and Yamahuchi, 2009) Natural contribution to changes in land cover includes storm, erosion, flood, earthquake and sea level rise What is mentioning is that human activities are intensifying these natural impacts on land cover change In recent years, growing human population has led to dramatically changes in land cover Moreover, land cover change is supposed as a primary component of many current environmental concerns (Frimpong, 2011) In other words, land cover change is gaining recognized a key driver of environmental change Changes in land cover and land use are pervasive and rapidly increasing which can have adverse impacts at local, regional and global scales (Frimpong, 2011) Therefore, qualitative and quantitative information about land cover changes are useful for sustainable management of natural resources, monitoring environment change, and minimize adverse consequence of environment as well Understanding land cover change can provide essential information for forming policies regarding social economic development and environment Therefore, accurate and update information on land cover change is necessary for building a suitable future land use policy

Satellite data with their repetitive nature have proved to be useful in mapping land use/land cover patterns and temporal changes Multi-temporal imagery data can be used for detecting and quantifying land cover change over time, thus providing significant information for monitoring environmental change and natural resource use (Bottomley, 1998)

In addition, the integration Remote Sensing technology and Geographical Position System (GIS) in assessing the changes in land use/land covers is used increasingly due to its powerful - cost effectiveness Remote sensing based techniques are now able provides vital tools for the assessment of land cover/land use in the management, with provide valuable and timely information on the nature resources and the environment GIS can provide a flexible environment for collecting, storing, displaying and analyzing digital data necessary for change detection (Reis, 2008) By using these techniques, it is possible to obtain a series of synoptic data for large geographic areas uniformly in time and space without exhaustive and expensive field surveys

Coastal wetland areas are highly vulnerable by both human activities and natural calamities With more than 3200 km of coastline, Vietnam has a large area of coastal wetland that provides a diverse range of natural resource and favorable conditions for social and economic development (fisheries, aquaculture, agriculture, tourism, transportation, urbanization, etc.) Notably, eighty percent of Viet Nam's population is settled within wetlands (MARD, 2004) However, these coastal zones are also highly vulnerable by natural disaster such as storm, erosion, earthquakes, and sea level rise Moreover, human activities such as mangrove logging for shrimp ponds, fuel wood, irregular exploitation, are intensify these natural impacts Therefore, to hazard mitigation, environment protection, conservation and sustainable use natural resources, are concerns and purpose of management of these zone

Like others coastal wetland zones in Vietnam, Tien Hai Nature Reserve in Thai Binh province is very sensitive, vulnerable zone Covering 10km length of coastline, and located in the left bank of the Red River (estuary Ba Lat river), the Nature Reserve has been recognized by UNESCO, as a one of the core importance of Red River biosphere reserve in 2004 With located in estuary Ba Lat river, it has very high biodiversity, rich

aquatic, and high number of birds Having very rich natural resources, thus this area is a place has dense population, and associated with livelihood of local communities Therefore, there are no denial is that Tien Hai Nature Reserve play a vital role in socio-economic development, disaster prevention, environmental protection, and conservation and development of biodiversity It is worth noting that Tien Hai Nature Reserve is facing with overexploitation resources for daily consumption of local communities, and exportation Natural resource has been depleted by overexploitation, unsustainable use resources of local communities In addition, the area is place that is always incurred threats caused by climate change such as natural disaster, sea level rise and environment pollution Tien Hai Nature Reserve is typically place about alternative dramatically of land cover due

to strong impact from human activities and natural calamities In recent years, with global climate change, and sea level rise is more concentrated, a threatening for coastal zones Hence, if management plan is lax, unreasonable or inconsistent policy and legislative system for coastal wetland management, that will lead to degraded of zone, environment change, loss of biodiversity These consequences will effect directly on human, and threaten human security

Recognizing the vulnerability of the region along with significance of monitoring land cover in management resource; as well advances of GIS and RS for the management,

so the topic "Monitoring land cover changes in Tien Hai Nature Reserve, Thai Binh province using multi-temporal Landsat data" is selected This study used a series of

Landsat satellite images from 2001 to 2015 to analyze land cover change, thus providing

an overview and updating information about coastal land cover and its change the last 14 years

Chapter II LITERATURE REVIEW 2.1 Applications of GIS and Remote Sensing to monitor land use and land cover

2.1.1 Land use and land cover

Land use include the modification and/or management of land for agriculture, settlements, forestry and other uses including those that exclude humans from land (Ellis et

al 2009)

Land cover refers to the physical and biological cover over the surface of land, including water, vegetation, bare soil, and/or artificial structures (Ellis et al 2009) Land cover is characteristics of land that can be observe physically, as by remote sensing

2.1.2 GIS - Remote Sensing and monitoring land use - land cover

Remote Sensing is a process of acquiring information about the Earth's surface without actually being in contact with it This is done by sensing and recording reflected or emitted energy and processing, analyzing and applying that information.(Canada Centre for Remote Sensing, 2008)

Geographic Information System (GIS) is a system designed to capture, store, manipulate, analyze, manage, and present all types of geographical data

Integration of GIS and remote sensing technologies is effective tool for management of natural resource When land use and land cover change became a central component in current strategies for managing natural resources and monitoring environmental changes Remote Sensing technique for land cover change detection and monitoring in recent times has been used to assess the change in land cover two or more time periods caused by environmental conditions and human actions The integration of Remote Sensing and GIS approaches, are providing new tools for advanced environmental

local, regional, and global scales The quantification of land use land cover changes is very useful for environmental management, policy makers and for public to better understand the surrounding

2 S tu s o f a lic a f G IS a R e m o S e s in c t la s e c o v e r

Key world studies of coastal land use and land cover:

In recent years, we have facing with growing population and climate change and; the issues about environment and natural resources are paid to great concern worldwide Moreover, with strong development of modern technology, it is recognized the significant role of LULC in management of natural resource and environment Therefore, leading to more and more number of studies of LULC were carried out by researchers, scientists, managers; especially in current changing climate There are several researches in the world:

In the study of AL Sghair (2013), author used remote sensing (Landsat TM and aerial photographs) and GIS technical combination with ground truthing work to assess wetland vegetation change over time at two contrasting wetland sites including freshwater wetland and saltmarsh in the UK Unsupervised and supervised classifications were used in this study to interpret the change of vegetation cover The author had showed the temporal change in vegetation during the period of study, most likely primarily produced by the management programme, rather than being due to natural change In his study, he proved that the RS and GIS approach can provide useful baseline data about wetland vegetation change over time, and across quite expansive areas, which can therefore provide valuable information to aid the management and conservation of wetland habitats

Srivastava et al., (2014) conducted; the study of monitoring land cover changes in coastal tract of Odisha using Landsat ETM+ imagery The land cover change dynamics were investigated by integration Remote Sensing approach and GIS The images were

classified using unsupervised classification and a post-classification comparison approach was used to detect the changes The result of study was indicated that an increase in agricultural areas while there is decrease in water bodies, soil and bare areas

In short, the integration of GIS and RS in the world was appeared from early and obtained many achievements in monitoring natural resource and environment The researchers used satellite image such as Landsat, Spot, Aster, Modis; and choosing image will depend on each different objects Sometime, to have process of interpretation with high accuracy, the some authors used combined with aerial photography with high resolution In the study, the authors also carried out many different classification methods

to compare result of each methods such as unsupervised, supervised, NDVI In generally, using GIS and Remote sensing in the world were fairly diversified, both material data and method

Key studies of coastal land use and land cover in Vietnam:

In Vietnam, there are many studies and researches using RS and GIS to monitor LULC, construct map; and detect the forest cover change, which support considerably to management of natural resources and environment

In the study of Nguyen Hai Hoa at el., (2013), authors assessed spatial-temporal changes in the extent and width of fringe mangroves, and changes in adjacent land use The study focused on the Kien Giang Province, Vietnam, for the period 1989-2009, and used four Landsat TM images (1989, 1992, 2003, and 2006) and three SPOT images (1995, 2003, and 2009) Supervised classification was conducted for image classification and mapping changes using a Maximum Livelihood Classifier approach The result of assessment of Landsat TM images showed a significant decrease in mangrove extent for the periods 1989-1992 (-2.7% yr -1) and 2003-2006 (-2.1% yr-1) SPOT images were used

to classify shoreline land cover into four types, namely mangroves; shrimp farms; built-up

areas; and bare and wet ground areas The loss of fringe mangrove extent was explained by the increase in extent of other land covers, especially shrimp farms and bare and wet ground, during the period of 2003-2009 The overall accuracy assessment of the classification using 2003 and 2009, SPOT 5 images were 90.1% and 90.5% respectively The overall accuracy assessment of the 2003 Landsat image was 92.1% The accuracy assessment showed that use of Landsat images was adequate for assessing spatial-temporal mangrove dynamics when higher spatial resolution satellite images were limited

The study was conducted by Tran Trong Duc (2003) with the focus of monitoring mangrove forest change in Can Gio district from 1993 to 2003 The author used post-classification comparison method to monitor mangrove forest change over time by using; Landsat and Aster data The map of mangrove forest is built with overall accuracy 80% The finding showed that there is a decline of mangrove forest area from 39000 ha in 1993

to 36000 ha in 2003 The main reason led to the change of mangrove forest in Can Gio district due to inappropriate policy management in relation to coastal management

Hanh Tran at el., (2015) conducted the study have aim to assess the spatio-temporal dynamics of land cover/land use changes in the lower Mekong Delta over the last 40 years from 1973 to 2011 with the coastal Tran Van Thoi District of Ca Mau Province The moderate to high spatial resolution (Landsat and SPOT) satellite images were used for analyze land use/land cover change, with 6 intervals of time Seven land cover class were identified in the site, and classified in the ENVI 5.0 and ArcGIS 10.0 (ESRI) software Classification of map given accuracy with the lowest is 81%; and the highest is 89% From

1973 to 2011, bare lands, cultivated lands, mangrove forest, and melaleuca forest decreased In contrast, aquaculture lands and built up areas increased The research also identifies the main drivers of land use changes in this delta region, which include economical policies as well as demographic, socio-economic, and environmental changes

In short, the integration of GIS and Remote Sensing in monitoring and management

of natural resource in Viet Nam was used widely, with more different field There were many kinds of Remote Sensing data were used in the researches such as Landsat, SPOT, Aster The ArcGIS and ENVI software were applied in image processing and spatial analysis The classification result given accuracy more than 80% The studies also identified main drivers of land use and land cover change; polices were majority reason leading to this change In generally, all of studies achieved goals which purpose

2.2 Studies of land use and land cover in Tien Hai Nature Reserve

2.2.1 Key study

The studies of land use and land cover in Tien Hai Nature Reserve are very little There is a research of Nguyen Thanh Hien at el., (2004), the study assess the change in land cover relate to shrimp ponds in two districts of northern Vietnam, namely Giao Thuy and Tien Hai Images from the Landsat TM and ETM+ sensors were used in this study from 1986 to 2001 This study aimed to detect the changes by using two hybrid change detection methods applied on TM/ETM+ images, and elucidated deforestation's causes in two districts through different trends of mangrove alterations The result showed that 63%

of mangrove areas have been replaced by shrimp ponds in Giao Thuy district from 1986 to

2001 And from 1994 to 2001 in Tien Hai, there are approximately 70% of its mangroves; most of them were replaced by aquaculture ponds Group of author was indicated that the shrimp farming have potential negative impacts on environment, one of which is wetland deterioration The lost of mangroves, so popular in this region, leads to the loss of a natural protection against tropical typhoons which are extremely important for the local population’s livelihood and paddy production

2.2.2 Key significant issues in Tien Hai Nature Reserve

In recent years, with increasing of population and climate change, the land use and land cover in Tien Hai Nature Reserve is facing with dramatically change, threat to its natural resource as well as environment Besides, management board hasn't completed yet, lack of staff, technique, and instrument for management In addition, also lack of researches about monitoring coastal land use and cover change over series time while it is considered as a fundamental component in management plans of natural resource Thus, leading to weakness management, lack of specific management plans, and these solutions

to reduce factors cause damage for the area

Chapter III STUDY GOAL, SPECIFIC OBJECTIVES, METHODOLOGY

3.1 Study goal

Study goal: The primary aim of this study is to monitor land cover change in Tien

Hai Nature Reserve from 2001 to 2015 by using multi-temporal Landsat imagery data The findings of study will contribute to better management for Tien Hai Nature Reserves, Thai Binh province

Study hypothesis: In this study, the hypothesis of study is land cover changes have

been caused by human-induced drivers rather than natural-induced forces which is given

and will be tested by analyzing key drivers of land cover change in Tien Hai Nature Reserve during 14 years Moreover, the finding of research will provide better understanding of land cover changes and solution in coastal zone, which could then better inform coastal managers and policy makers aiming to lessen potential risks caused by natural induced and human induced

3.2 Specific objectives

To be consistent with study goal and hypothesis, the following objectives are

selected as below:

Objective 1: Investigate the status current of land cover and its management

schemes in Tien Hai Nature Reserve

Objective 2: To quantify land cover changes in Tien Hai Nature Reserve from 2001

to 2015 using multi-temporal Landsat data This objective includes two sub-objectives stated as below:

Construct a single map of land cover in Tien Hai Nature Reserve each year of

2001, 2006 and 2015

Monitor land cover changes in Tien Hai Nature Reserve in selected periods, namely 2001- 2006, and 2006 - 2015

Objective 3: To identify key drivers of land cover changes over the last 14 years in

Tien Hai Nature Reserve

Objective 4: To propose solutions for better management of coastal land use in Tien

Hai Nature Reserve

3.3 Study methodology

This part describes the methods and materials that were used and applied in the study, including data collection, data processing, data analysis with a view to achieve the set objectives First of all, flowchart below (Figure 3.1) will provide an overview about method proportion of each object

Figure 3.1: Overview flowchart of study objectives

Investigate the status current of land use land cover and its management schemes in Tien Hai Nature Reserve

Quantify land cover changes in Tien Hai Nature Reserve from 2001 to 2015

Identify key drivers of land cover changes over the study time in Tien Hai Nature Reserve

Propose solutions for better management of coastal land use in Tien Hai Nature Reserve

3.3.1 Data collection

3.3.1.1 Interview data

Interview data was collected from interviews which are considered as a good way

to get information directly from people live in there Due to they understand clearly about this study site, so collecting information from them is very true and useful for research From interview, we can know the status current of land cover, schemes or regimes of management, as well as driving factors of land cover change To interview has good result, survey questionnaire were designed (Appendix 1, 2) The interviews local people were carried out with 15 famers belong to three communes of buffer zone of the Nature Reserve; and number of farmers was chosen randomly during the processing of field work In addition, staffs of board management and local authorities of three communes also is one

of objects of interview

The management board of Tien Hai Nature Reserve

Interview staffs of management board to get information about the current land use-land cover of the area, as well as policy and regimes management, which are implemented in Tien Hai Nature Reserve Thence, could have an overview about the protected area, processing of formation and change in the area People belong to object of interview: staffs of board management in Resources and Environment department in Tien Hai district, local authorities of three communes: Nam Hung, Nam Phu, Nam Thinh

Local communities in three communes of buffer zone

Local communities play an important role in management of the Nature Reserve, because they affect directly on the Nature Reserve They understand better than anyone about the area Therefore, getting information from local people is way to have real and accurate those information about the area The interview local people also can have an overview about effectiveness of management plans of the Nature Reserve In this study,

the interview local communities to determine reasons drive change of land cover during 14 years, from 2001 to 2015 And local people who towards to object of interview includes old or middle-age people and people with experienced Due to, they could provide much information about the area in many years ago

3.3.1.2 Reference materials

Reference material is another method to get more information about the study site, besides interview data The study used some information from the peer-reviewed articles, report, news, newspaper, official documents about the Tien Hai Nature Reserve These findings can provide more information for investigate status of land cover, the management schemes of Tien Hai Nature Reserve in current, as well as reasons of change

in land cover in last time

3.3.2 Integration multi-temporal Landsat data and GIS

An integration of multi-temporal remote data and GIS provides a useful method to quantify coastal land cover change in Tien Hai Nature Reserve from 2001 to 2015 To have

an overview of land cover change in Tien Hai nature reserve during 14 years, the study

firstly constructed a single map of land cover types in Tien Hai Nature Reserve each year

of 2001, 2006, and 2015 to evaluate and comment status of land cover through each year Then monitoring land cover changes in Tien Hai Nature Reserve in selected periods, namely 2001- 2006, 2006 - 2015 Dividing certain period of coastal land cover is a good

way to detection more clearly about land cover change in given of period and how they changes The period of time is 5 and 9 years This part describes the study area, method and materials that were applied in the collection, processing, analysis data The flowchart below represents more detail sequences of steps

Boundary of study area

GPS Field survey

Landsat image

2001, 2006, 2015

Clip study area

Band combinations

Land cover change 2006-2015

Land cover change 2001-2015

3.3.2.1 Materials and data

3.3.2.1.1 Landsat images

Landsat satellites have been collecting images of the Earth's surface for more than forty years NASA launched the first Landsat satellite in 1972, called Landsat 1 and the most recent one, Landsat 8, on February 11, 2013

In this study, Landsat 5 TM imagery in 2006, Landsat 7 ETM+ imagery in 2001; and Landsat 8 OLI/TIRS imagery in 2015 were used These images are freely available from the Landsat archive in the United States Geological Survey (USGS)

This study site is coastal wetland zone, the feature of area is easily changed by natural factors such as season, tide Therefore, choice in satellite image is crucial for monitoring land cover in coastal wetland An image without cloud is obviously preferable But the date of acquisition images ensures fewer differences in solar illumination and atmospheric conditions between years In addition, the tide level is also a crucial factor in the choice of the satellite images for the detection of coastal changes Ideally, the tide level must be similar between the periods, but that does not always correspond with satellite passage Therefore, in this study tried to collect the satellite images with fewer difference

of date acquired (Table 3.1) All of the images are acquired in the rain season, so high soil

of moisture, and having approximately equivalent tide level

Table 3.1: Satellite images used for the study

Codes Satellite instrument Date of acquired Path x Row LE71260462001272SGS00 Landsat 7 ETM+ September 29, 2001 126 x 46 LT51260462006262BJC00 Landsat 5 TM September 19, 2001 126 x 46 LC81260462015191LGN00 Landsat 8 OLI/TIRS July 10, 2015 126 x 46

Source: earthexplorer.usgs.gov

Table 3.2: Ground control points (GCPs) and land cover classes

No Land cover classes Number of GCPs

3.3.2.2 Image pre-processing

All Landsat images used in this study were acquired under clear atmospheric correction The collected images were registered by standard projection, datum UTM 48 North, and the spatial resolution is 30 meters

3.3.2.2.1 Clipping study area

In this step, the boundary of the study site was created and used to cut the study area from the Landsat image using clipping tool in ArcGIS 10.1 software The boundary of the study area was formed as the shapefile and registered to the WGS84 UTM zone 48N

3.3.2.2.2 Band combinations

Each satellite download from USGS, with each band separated as a TIF file The processing to composite bands TIF files from the single band to many bands in a single image Landsat images are composed of many different bands with 7 bands to Landsat 5TM, 8 bands to Landsat 7ETM+ and 11 bands to Landsat 8, each representing a different proportion of the electromagnetic spectrum

Individual bands can be composited in Red, Green, Blue combination in order to visualize the data in color These combinations are made that will have different advantage using for classification each different feature such as forest, agricultural, building, water, Therefore changing these band combination and the RGB “channels” they are assigned to in order to highlight different properties of features in the image To distinguish between different cover types, these False Colour Combination (FCC) images are useful In order of this study, with five different land cover, I used FCC of RGB=6-5-4 for Landsat 8 image 2015 This band combination uses short-wave infrared channel (band 6), near-infrared channel (band 5) and red channel (band 4), help for clearly defined vegetations with green, high distinguishing between old forest and young forest, water is blue or black The bare soil, urban, agriculture area is also easy to identify This

combination prefers to vegetation and aquatic ecosystem analysis, due to color is natural to human eyes With Landsat 5TM and Landat 7 ETM+, I used FCC of RGB=5-4-3, this combination is similar to the combination 6-5-4 of Landsat 8 This combination is useful for vegetation studies, involves the red channel (band 3), the near infrared channel (band 4), and the mid-infrared channel (band 5) The combination provides the user with a great amount of information and color contrast Healthy vegetation is bright green and soils are mauve Figure 3.3 represents the band combination of Landsat images in 2001, 2006, and

In this study, in order to classify image, five land cover classes were chosen to identify, including water, built-up and sand, casuarinas (other coastal plants), mangroves, and others (tidal plat, salt mud, bare soil, non-identified) Description of these land cover classes are represented in Table 3.3

Table 3.3: Description of land cover classification used

Land cover

classes

Description

Sample of satellite image

Water

Permanent open water, sea, river, lake, pond,

gully; tidal flat

Built-up and

sand

Settlements, roads, and any other kind of

infrastructure Sand dune, coastal sand

Casuarinas Casuarinas in dike edge, road side, and coast

There are two major approaches to classifying the pixels in a multiband raster:

supervised and unsupervised classification In this study the Iso cluster unsupervised classification to identify the land cover classes is used by using the Image Classification

tool in ArcGIS Unsupervised classification is a method of identifying, grouping, and

labeling features in an image according to their spectral values In unsupervised classification, pixels are clustered together based on spectral homogeneity and spectral distance The advantage of unsupervised classification is that it is automated, inexpensive, and does not require a priori knowledge of the study area, can be easily be conducted for large areas This makes it can easily be conducted for large areas and needing less skills and experience

In the method of unsupervised classification, itrequires setting number of classes, minimum of class size and interval sample Number of spectral classes we set will be dependent on the number of informational classes, the complexity of your area of interest and the final project With five classes of land cover, the study selected 30 for number of classes and after running unsupervised classification, computer display the findings with

30 unique values Then, the study needs to determine which informational class contains each one of these spectral classes andultimately reclassify all 30 spectral classes into five classes of land cover

To help for deciding what features present and belong to specific informational classes, this study used GCPs from field survey by using GPS, from Google Earth and other status of mangrove maps in study areas In process of classification image, due to spectral value of built-up and sand classes is equivalent, so it difficult to separate it into two different classes, therefore I added up them into one classes

3.3.2.4 Accuracy assessment

Accuracy assessment is an important part of any classification With that, the study could know how accurate the classification is, by using GCPs from field survey and the Google Earth in 2015 The GCPs are used for land cover analysis in 2015 With 660 GCPs, 20% of 660 points (equivalent to 132 points) were randomly selected the study and used to assess the accuracy of image classification In this accuracy assessment, the study used

error matrix to represent accuracy of classification, which is a square array of rows and columns, presenting the relationship between the classes in the classified and field survey

by GPS In overall accuracy, producer's accuracy and user's accuracy were derived from error matrix

5.2.5 Change detection

This step is used to detect land cover change during the study period of time The

study used Raster Calculator Tool (ArcToolBox) to calculate the cells of two maps

together, so the output will indicated how change is detected for each feature class each period From change detection, the findings will provide overview information about the change of land cover through periods and a final map of land cover change over periods

Chapter IV STUDY SITE, NATURAL AND SOCIO-ECONOMIC FEATURES

beach in Nam Phu commune (Source: birdlife international)

Together with the Xuan Thuy National Park in Nam Dinh province – the first recognised site in Southeast Asia, the Tien Hai Nature Reserve in Thai Binh province plays

Ramsar-a significRamsar-ant role in preserving nRamsar-ationRamsar-al Ramsar-and internRamsar-ationRamsar-al biodiversity The site is one of core zones of Red River Delta Biosphere Reserve, and was recognized by UNESCO as a world biosphere reserve in 2004 According to IUCN (International Union for Conservation of Nature) protected area categories system, Tien Hai Nature Reserve is

classified as habitat/species management area (category IV), with the aim to protect particular species or habitats and management reflects this priority Moreover, according to IBA criteria (Important Bird and Biodiversity Areas), the Nature Reserve are classified as A1 (globally threatened species)

Climate

The Tien Hai Nature Reserve has a monsoon humid tropical climate The average temperature is 150C in winter and 380C in summer The rain season starts from May to October and between October and January of the following year is dry season The average annual rainfall from 1600 to 1800mm, and approximately 80% of total rainfall appeared in rain season

Hydrology and tide

The Tien Hai Nature Reserve was supplied water and alluvium from Red river There are two major rivers in the nature reserve: Lan and Hong rivers Beside, two main river, there are many steams from Hong river drain through mangrove forest to provide water and alluvium for growing of mangrove forest Every year, Red river supplied large amount of alluvium and sediment for the nature reserve, average 6-7kg/cm3 in the rain season The average amount of alluvium and sediment from Red river flow the sea about

114 million tons (Dr Do Van Nhuong and Mr Hoang Ngoc Khac 2001) Tide of Tien Hai Nature Reserve is diurnal tide The tide is relative weak, about 150-180 cm per day, with 3,3-3,9 m to max tidal water level, 0.4 m to min tidal water level Due to tide and hydrology created condition for growing of mangrove forest, aquaculture production, typically raising platform land and expand area for the Nature Reserve The main types of soil in the Tien Hai Nature Reserve include sandy soil, salty soil, saltmud soil and alluvial soil

Biodiversity

The Natural Reserve is rich in mangroves and casuarinas forests, shrimp ponds, tidal flat and sand dunes and banks It is home to 215 species of birds and Platalea minor (cò mỏ thìa) is a species in Red Book In addition, along with 115 species of higher plants – a major source of food for birds, 43 medicinal plants, 113 insect species, 107 fish species and 37 amphibians and reptile species Moreover, the mangrove forest is important ecosystem of the Nature Reserve, with diversity of mangroves with majority species such

as Kandelia candel (trang), Aegiceras corniculatum (sú), Sonneratia caseolaris (bần chua), Acanthus ilicidfolius (Ô rô), Deris trifoliata (Cóc kèn), Avicennia marina (Mắm biển) In addition, Causuarina forest in the seashore and dyke are planted popularly in this zone

4.3 Socio-economic features

The population of three communes in buffer zone of the Tien Hai Nature Reserve is

16014 people, including 6620 people in Nam Thinh, 5372 people in Nam Hung; and 4635

people in Nam Phu (Socio-economic newspaper Tien Hai district, 2008) Almost people

are a famer, and large different in income of local people

Inside the Nature Reserve, local communities are engaged in a number of economic activities, including aquaculture, livestock raising, fishing and collection of shellfish The principal species being collected were Lingula sp (giá), Glauconome chinensis (don), Meretrix (vạng), Mactra quadrangularis (ngao bốn cạnh) and Cyclina sinensis (ngó đỏ) The average daily harvest was estimated at 1.9 tones, with an estimated value of US$529 (Pedersen and Nguyen Huy Thang, 1996) In generally, main activities of Tien Hai Nature Reserve include aquaculture production, exploitation fishery, forestry and tourism

Because the Tien Hai Nature Reserve extend area to the sea In this study, I narrowed the area of monitoring land cover to easier for field survey The study site cover 7305.8 ha, is also bordered by the Red River (also called Ba Lat River) to the south, the

Lan river to the north and the main sea dike to the west belong three communes, and sea to the east The study site narrowed area to the sea compare to the totally area of Tien Hai Nature Reserve, (Figure 4.1)

Figure 4.1: Location of study site in Tien Hai district

Chapter V RESULTS AND DISCUSSION 5.1 Current status of coastal land cover and its management schemes

5.1.1 Current status of coastal land cover in Tien Hai Nature Reserve

As findings indicated, Tien Hai is a wetland zone, with 12500 ha of area Currently, the Tien Hai Nature Reserve has 1400 ha of coastal mangroves, 100 ha of casuarinas, others area covered by wetland, mudflat and sand dune With 1400 ha of coastal mangrove area spatially distribute in Nam Hung with 600 ha, Nam Phu with 350 ha, and Nam Thinh with 450 ha (according to Mr Pham Truong Son - Deputy Head of management)

Interestingly, coastal mangroves in study site distribute mainly in two sides and branches of Red River (known as the Hong river), which is found with the rich alluvium content and be very appropriate for mangroves growing In the Nature Reserve, coastal mangroves are found with mixed forest, including from 2 to 4 mangrove species with the age more than 5 years old, namely: Kandelia candel (trang), Aegiceras corniculatum (sú), Sonneratia caseolaris (bần chua), Acanthus ilicidfolius (Ô rô) In addition, there are small areas of mangroves found distributing sparsely in aquacultural ponds as an ecology aquaculture model accounting for about 8% of all mangrove forested area A mixed mangroves and shrimp pond (ecological aquaculture model) are found as the most common model in the Nature Reserve The density coverage in aquacultural ponds was identified about 30% of areas which includes mangroves, grass/shrubs and other plants (Tran Tan Dat - Director of community education center in Nam Phu commune)

Coastal zones and dyke edge are covered by casuarinas trees with purpose of protection seashore and sea dyke, preventing sea waves and sand from drift As a coastal wetland zone, the Tien Hai Nature Reserve is covered mainly by mudflats A large area of tidal mudflat located in the Northern part of the Nature Reserve, and belong Nam Thinh

commune, covering 1500 ha of area This zone has a vital role in livelihood of local communities, using with purpose bivalves production (Mr Bui Kien Quyet - Chairman of Nam Thinh commune)

5.1.2 Management schemes of Tien Hai Nature Reserve

In general, Tien Hai Nature Reserve has not had a specific Management Board yet People's committee of Tien Hai district is administrative authorities of the Nature Reserve

In 2001, Thai Binh province approved reestablishment Management Board belongs to People's committee of Tien Hai district The Management Board of Nature Reserve has three staffs, including one Head of management board, one Deputy Head of management board and one officer In addition, the management of the area is also assigned for three communes of buffer zone, including: Nam Hung, Nam Phu and Nam Thinh

All of plans and projects that Management Board proposed and get approved by People's committee of Tien Hai district; and will be implemented by each communes Each commune has responsibility for their management; and plan implementation in their given administrative area Mangrove forest in Tien Hai Nature Reserve is allocated for each communes directly management Each commune established a protection group including

5 - 6 people They have given a task of protecting forest, preventing illegal encroachment from local people on the area They operate based on grant in project 661 of government The establishment of protection group has a positive contribution in management and protected forest in commune In addition, reforestation is promoting in three communes According to Mr Pham Truong Son - Deputy Head of Management Board, management plans of the Tien Hai Nature Reserve currently always combine with communities, based

on communities, and opinions of communities are taken into plans From institutional rules, these management plans are implemented effectively and get involved from local participations The management plans of the area increased effectively, as a result of