Assessing cropping pattern adaptability to climate risks in the vietnamese mekong river delta

VIETNAM NATIONAL UNIVERSITY, HANOI VIETNAM JAPAN UNIVERSITY TRAN THUY TRANG ASSESSING CROPPING PATTERN ADAPTABILITY TO CLIMATE RISKS IN THE VIETNAMESE MEKONG RIVER DELTA MASTER'S T

VIETNAM NATIONAL UNIVERSITY, HANOI

VIETNAM JAPAN UNIVERSITY

TRAN THUY TRANG

ASSESSING CROPPING PATTERN

ADAPTABILITY TO CLIMATE RISKS

IN THE VIETNAMESE MEKONG

RIVER DELTA

MASTER'S THESIS

VIETNAM NATIONAL UNIVERSITY, HANOI

VIETNAM JAPAN UNIVERSITY

TRAN THUY TRANG

ASSESSING CROPPING PATTERN

ADAPTABILITY TO CLIMATE RISKS

IN THE VIETNAMESE MEKONG

PLEDGE

I pledge that this thesis is original and has not been published before I am aware of the regulations regarding using other research and documents, and I will ensure that all citations and references adhere to the requirements

I have reviewed the guidelines on plagiarism violations I solemnly declare that the research presented in this thesis is my own and does not infringe upon the Regulation

on Prevention of Plagiarism in Academic and Scientific Research Activities at VNU Vietnam Japan University (Issued together with Decision No 700/QD-ĐHVN dated 30/9/2021 by the Rector of Vietnam Japan University)

Author of the thesis

Tran Thuy Trang

I am indebted to Dr Nguyen Thi Thuy Hang for her technical expertise and professional advice Her insightful feedback and constructive criticism have contributed significantly to the improvement of this thesis

I would like to acknowledge my classmates, who have been an incredible source of support throughout this journey Their camaraderie, willingness to help and mutual care for one another have created a positive and conducive learning environment Late-night study sessions and shared deadlines have brought us closer together, and I

am thankful for their companionship

Lastly, I want to acknowledge the numerous challenges and setbacks I encountered during this research It is through perseverance and determination that I was able to overcome these obstacles I commend myself for not giving up on life and staying committed to the completion of this thesis

Without the support and contributions of these individuals and my personal resolve, this thesis would not have been possible I am sincerely grateful to all who have been a part of this journey and have contributed to its success

TABLE OF CONTENT

PLEDGE i

ACKNOWLEDGEMENT ii

TABLE OF CONTENT i

LIST OF TABLES i

LIST OF FIGURES ii

LIST OF ACRONYMS iv

CHAPTER 1 INTRODUCTION 1

1.1 The necessity of the research 1

1.2 Literature review 3

1.2.1 Climate-risks in Vietnamese Mekong River Delta 3

1.2.2 Cropping pattern in Vietnamese Mekong River Delta 7

1.2.3 Sustainable cropping methods in Vietnamese Mekong River Delta 12

1.3 Scope of the research 15

1.4 Research questions and hypotheses 15

1.5 Research objectives 16

1.6 Study area 17

1.7 The framework of the research 23

CHAPTER 2 DATA AND METHODS 25

2.1 Data used 25

2.2 Methods 25

2.2.1 Remote sensing data collection and analysis 25

2.2.2 Ground truth observation 30

2.2.3 Desk review method 37

2.2.4 Non-structured interview 37

CHAPTER 3 RESULTS 39

3.1 Cropping pattern changes in Vietnamese Mekong River Delta 39

3.2 Cropping pattern changes adapting to climate risks 43

3.2.1 Cropping pattern changes adapting to floods 44

3.2.2 Cropping pattern changes adapting to other climate risks 48

CHAPTER 4 DISCUSSIONS AND RECOMMENDATIONS 54

4.1 Assessing the cropping pattern changes’ efficiency 54

4.2 Problems with triple cropping pattern 55

4.3 Recommendations 56

CHAPTER 5 CONCLUSION 60

5.1 Conclusion 60

5.2 Limitations and future outlooks 61

REFERENCES 63

APPENDICES 68

i

LIST OF TABLES

Table 1-1: Comparison between costs and benefits of intensified and balanced

cropping in An Giang and Dong Thap, respectively 13

Table 1-2: Explanation for sustainable cropping practice approaches in VMRD 14

Table 1-3:Researching questions and hypotheses 15

Table 1-4: Approved agriculture development plan in VMRD period of

2011-2025 20

Table 2-1: IGBP classification scheme and RGB color code for LULC map 26

Table 2-2: MODIS data acquisition date table 27

Table 2-3: GTO field trip schedule in VMRD (March 2023) 31

Table 2-4: EVI accuracy assessment of Tra Vinh province points 33

Table 2-5: Accuracy assessment of An Giang province points 34

Table 2-6: Heading dates accuracy assessment of Tra Vinh province points 35

Table 2-7: Heading dates accuracy assessment of An Giang province points 36

ii

LIST OF FIGURES

Figure 1-1: Flooding period by percentage of area in VMRD from 2000 to 2022

(GIS data) 5

Figure 1-2: Provincial flood and salinity risk in VMRD 6

Figure 1-3: Typology and short definitions of the cropping system components 8

Figure 1-4:Dikes and planted area of crops in An Giang from 1985-2016 11

Figure 1-5: Dike construction: low and high dikes area in An Giang province in 2011 and 2014 12

Figure 1-6: VMRD administrative map 18

Figure 1-7: 3 ecological sub-regions for agricultural development plan in VMRD 20 Figure 1-8: SWI leading to soil salinity map with the depth from 0 cm to 20 cm in Tra Vinh province, VMRD, VN 22

Figure 1-9: An Giang and Tra Vinh in administrative map 23

Figure 1-10: Dike area in An Giang Province, VN, in 2014 23

Figure 1-11: The logical framework of the research 24

Figure 2-1: EVI spectral viewer 29

Figure 2-2: GTO in VMRD route (blue line) 31

Figure 2-3: GTO validation process from 1 point 32

Figure 2-4: GTO route in Tra Vinh and An Giang (pictures for each location are attached and can be accessed via Google Earth) 33

Figure 2-5: The example of rice growth stages and other rice field conditions in the paddy field area 34

Figure 3-1: LULC maps from 2001 to 2021 in VMRD 39

Figure 3-2: LULC change in VMRD from 2001 to 2021 (GIS data) 40

Figure 3-3: Differences in planted area of paddy by provinces in the whole period from 2001 to 2021 (statistical data) Error! Bookmark not defined. Figure 3-4: Map of changes in planted area of paddy in VMRD by provinces from 2001 to 2021 (statistical data) 41

Figure 3-5: Cropping frequency changes in VMRD from 2000 to 2022 43

Figure 3-6: Yearly cropping frequency in VMRD from 2000 to 2022 (GIS data) 43

Figure 3-7: Maps of the number of flooding days in An Giang province from 2008 to 2015 45

Figure 3-8: Maps of cropping frequency in An Giang province from 2008 to 2015 46 Figure 3-9: Cropping frequency changing trend in An Giang province from 2010 to 2015 (GIS data) 47

Figure 3-10: Production of paddy in An Giang from 2010 to 2015 (statistical data) 48

Figure 3-11: LULC maps in Tra Vinh from 2001 to 2021 49

iii

Figure 3-12: Number of inundated days yearly in Tra Vinh and An Giang from

2000 to 2022 50Figure 3-13: Planted area for paddy in Tra Vinh province from 2001 to 2021 (statistical data) 51Figure 3-14: Cropping frequency in Tra Vinh province from 2000 to 2022 (GIS data) 52Figure 3-15: Cropping frequency maps in Tra Vinh province from 2000 to 2022 52Figure 4-1: Yield of spring, autumn & winter paddy by provinces in VMRD from

2001 to 2014 (statistical data) 54Figure 4-2: Paddy and orange intercropping in Tra Vinh Province (GTO) 57Figure 4-3: Lotus pond combined with paddy field in Long An Province (GTO) 57Figure 4-4: Inefficient paddy fields transformed into land for installing solar power panels in An Giang Province (GTO) 58Figure 4-5: Advertisement material for sustainable agriculture approach in VMRD:

3 reductions 3 gains (left), and 1 must-do 5 reductions (right) 59

iv

LIST OF ACRONYMS

1

CHAPTER 1 INTRODUCTION

1.1 The necessity of the research

Vietnam (VN) is located on the East Sea of the Pacific Ocean with a long coastal line

at more than 3,000 km of coastline and is among the most vulnerable countries to the impact of climate change (CC) In reality, VN has been through many hostile phenomena of CC in recent years, including sea level rise, temperature increase, and intensified and more frequent hydro-climatic disasters (Cruz, 2007) This can be attributed to the country's location, long coastal line, and complex hydrological system Provided that the economy relies heavily on agriculture, this sector accounts for over 12% of the national GDP and 24 million employment (Nguyen et al., 2022)

In 2021, the rice production in VN was approximately 43.9 million metric tons, and it

is one of the biggest rice exporters in the world (Nguyen & Scrimgeour, 2022) Accounting for more than 50% of the nation's rice yield (Nguyen et al., 2022), the Vietnamese Mekong River Delta (VMRD) is one of the deltas cultivated most intensively worldwide

However, this region is vulnerable to climate change and disasters because of its lying landform According to Intergovernmental Panel on Climate Change (IPCC) report, the climate risks for agriculture are rising temperature, changing rainfall, acidification increase, lack of oxygen, sea-level rise, extreme events, increased storms, and cyclones, droughts, and floods, increased climate variability (Intergovernmental Panel on Climate Change (IPCC), 2014) The MRD also faces several climate risks, including rising sea levels and exacerbating flooding phenomena, which also lead to saltwater intrusion (SWI) in MRD, along with droughts and limited availability of freshwater

low-The VMRD is most affected by flooding and inundation low-The harvest of rice is reasonably susceptible In a report from (Nguyen et al., 2007), local farmers blame either inadequate flood control systems or damaged crops for low yields or crop losses and claim that flooding has practically become an annual occurrence Many places

2

along the central and north-central coasts are prone to flooding with strong currents These pose a threat to property, human life, irrigation systems, and public infrastructure and are made worse by dyke breaches, strong winds, and sea waves Farmers with homes or farms close to the seaside are at risk from storm-caused sea surges (Nguyen et al., 2007) Another significant problem in VMRD is SWI The SWI issue stems from the Mekong River's tidal level is lower than the high tide at sea, causing the river's flow to invert with the tides and bring water inland (Thuy & Anh, 2015) Salinity and erosion issues would also result from this Sea water travels 70 kilometers inland during the dry season (Noh et al., 2013) This problem can potentially alter the spatial organization or configuration of the landscape because salinity can change the properties of water and soil components of the pattern, corrupt the flow of nutrients, and negatively affect rice farming in the area because rice has a low salinity tolerance

In the VMRD, a dynamic and interdependent relationship exists among climate risks, land use and land cover (LULC) change, and cropping pattern changes These factors are intricately linked and influence one another in complex ways Climate risks, such

as floods and SWI, can significantly impact LULC change in the region Changes in LULC, such as land erosion from the sea-level rise or conversion of agricultural land, can, in turn, affect the local climate and exacerbate climate risks Additionally, changes in cropping patterns, such as shifts from traditional rice cultivation to other crops or cropping systems, can be influenced by both climate risks and land-use changes These changes in cropping patterns can have implications for agricultural productivity, food security, and the overall sustainability of the VMRD Therefore, understanding and managing the dynamic interactions among climate risks, LULC change, and cropping pattern changes is crucial for sustainable development and adaptation strategies in the VMRD

With the importance and vulnerability of MRD's agriculture, it is necessary to understand the climate risks and how to cope with them by more resilient cropping practices As we know, the cropping pattern and the climate are intimately related, as one can reflect the changes of the other Therefore, the goal is to understand the

3

region's cropping patterns fluctuation by using the time-series imagery observation method Analyzing the data will enlighten the underlying reasons for the cropping pattern changes throughout the years and the connection to the local climate risks, thereby seeking and endorsing more resilient and sustainable cropping practices for each area

1.2 Literature review

1.2.1 Climate-risks in Vietnamese Mekong River Delta

Climate risks are associated with the potential adverse effects of climate change on the natural environment and human society, encompassing sudden events such as floods, droughts, and storms as well as gradual shifts in temperature and precipitation patterns and sea-level rise over a more extended period (IPCC, 2014) including effects on,

agriculture, water resources, infrastructure, and public health (Climate Change Impacts

in the United States: The Third National Climate Assessment, 2014) The World Bank

conducted a study in 2019 titled "Climate Change and Migration in Coastal Vietnam," which identified floods, droughts, and sea-level rise as the most prominent climate

risks in VMRD According to a report by the (Climate Change Vulnerability and Adaptation Assessment of the Mekong Delta, 2016), the Mekong Delta in VN is

vulnerable to a range of climate risks These include floods, droughts, SWI, and land erosion, which can cause significant harm to the region's infrastructure, economy, and public health These risks also threaten the livelihoods of the communities living in the area

This research will focus on the following climate risks: floods, droughts, and SWI

- Floods: Floods are considered a climate risk in the Vietnamese Mekong River Delta (VMRD) due to the region's low-lying topography, which makes it highly susceptible to flooding during the rainy season Climate change is anticipated to worsen this issue by amplifying the occurrence and severity of extreme weather events, including floods The region has experienced more frequent and severe floods

in recent years, with a severe flood in 2011 with a return period of between 10 and 20

years (Mekong Delta Flood Report, n.d.) According to the Mekong River

- SWI: SWI is a major climate risk in the VMRD, particularly during the dry season when water levels in the river are low SWI in the VMRD is caused by a combination of factors, including sea-level rise, reduced river flow, and land subsidence The VMRD has been increasing in recent years, affecting a large part of agricultural land in the region SWI in VMRD has numerous impacts beyond just agriculture It also has social and economic consequences, affecting public health and infrastructure SWI in the region can also lead to the degradation of soil quality, which can cause low agricultural productivity and income for farmers It can also affect drinking water quality and cause health problems, particularly for kidney disease patients Furthermore, saltwater can damage roads, bridges, and other infrastructure, affecting transportation and economic activities

5

Figure 1-1: Flooding period by percentage of area in VMRD from 2000 to 2022 (GIS

data) Generally, various climate data sources exist, including meteorological stations, satellite observations, climate models, and research studies These sources provide valuable information on temperature, rainfall, sea-level rise, and other climate-related factors In recent years, increased attention and research have focused on climate risks

in the VMRD Efforts have been made to improve the region's data collection, analysis, and mapping of climate risks

For mapping climate risks on VMRD, the Vietnamese government has made significant efforts in creating the CS-MAP (Climate-Smart Mapping and Planning) CS-MAP focuses on mapping and analyzing climate change risks and vulnerabilities and identifying suitable adaptation and mitigation strategies for sustainable development There is also a participatory approach in creating CS-MAP where stakeholders from multiple fields (Department of Agriculture and Rural Development, Department of Crop Production, national research institutes, Crop Production and Plant Protection Office and Hydrological Management Office) and different administrative levels (province, district, commune, village) participated in the discussion (Yen et al., 2019)

Wassmann et al (2019) conducted one of the most remarkable studies to create a resolution mapping for floods and SWI in VMRD to help farmers decide on rice

NUMBER OF FLOODING DAYS

Flooding period by percentage of area in VMRD from

2000 to 2022

7

policies in both regional and local levels; (5) restraints in resources (funding and technical capacities) to conduct further research and carry out the action plan

1.2.2 Cropping pattern in Vietnamese Mekong River Delta

1.2.2.1 Definition of cropping pattern

An overview of the cropping situation in VMRD can be found in the report from the (Food and Agriculture Organization (FAO) of the United Nations (2020), which stated that the cropping pattern in the VMRD is dominated by rice production, which covers about 70% of the total cropland Other crops such as maize, sweet potato, cassava, and vegetables are also cultivated but on a smaller scale The report highlights that there has been a shift in cropping patterns in recent years, with a decrease in the area dedicated to rice and an increase in the area for other crops This is attributed to changing market demands, increased mechanization and irrigation, and the impacts of climate change on agriculture

Among many different definitions of cropping pattern, Bégué et al (2018)gave out a comprehensive model to explain the cropping system that is the premise for analyzing the cropping pattern (Figure 1-3)

in achieving high crop yields (Huynh et al., 2020) Therefore, depending on the climate condition of the year, the sewage release date, or the provincial regulation, the number of crops planted each year may vary And there three main cropping patterns are practiced in VMRD single cropping, defined as planting one crop per year; double cropping, planting two crops per year; and triple cropping, planting three crops in a year on the same plot of land The cropping calendar may vary between different places However, for the statistical report, there are three main cropping seasons in VMRD and VN in general: the spring season, which starts in November or December; the Autumn season starts in April; the Winter season starts in May or June According

to Thuy & Anh (2015), the rice production schedule in the Mekong River Delta varies among provinces and is influenced by hydrological regimes and irrigation systems

9

Farmers can plant three or four seasons of rice per year in areas with completed dike systems, even in flood-prone or coastal regions Conversely, there are usually only two rice seasons per year in areas lacking completed dike systems

1.2.2.2 Vulnerability of rice production in Vietnamese Mekong River Delta

As the concerns for the impacts of CC in VMRD increase, the number of studies on the vulnerability of rice production in VMRD also increases as this region is the biggest rice bowl of VN Due to the overall impacts of CC, the rice yield is predicted

to drop (Chun et al., 2016; Jiang et al., 2019) According to Jiang et al (2019), CC is projected to have a considerable impact on the annual rain-fed rice crop production in the entire VMRD, causing a substantial decrease of approximately 35% yearly compared to the current productivity levels in the region

According to a study by (Nguyen et al., 2007), flood and inundation is the most impactful risk in VMRD The study claims that flooding reduces crop yields and can even result in complete crop loss, affecting farmers' livelihoods in the region Floods also disrupt the timing of planting and harvesting, which can result in the loss of seedlings or harvests, and in some cases, farmers may need to replant or delay their planting schedules This can have long-term impacts on the cropping calendar, causing the need for adjustments for planting times and leading to a shift in cropping patterns

In addition to these direct impacts, flooding can also lead to secondary effects on cropping systems For example, floodwaters can cause soil erosion and sediment deposition, affecting soil fertility and making it difficult for crops to grow Floods can also increase the risk of crop diseases and pests, further impacting crop yields and quality

The findings from Thuy & Anh (2015)’s research on the climate change vulnerability

of rice production in VMRD indicate that coastal areas are more susceptible to extreme events due to their proximity to the sea The similarity in the distribution patterns of the two risks, flood, and SWI, suggests that without effective adaptation and mitigation strategies, the future increase in extreme events will have detrimental effects on rice production in the region, given the current vulnerable state of coastal areas The results of this research highlight that under CC, the paddy of VMRD will

for land use and strategic development (Mekong Delta Plan, 2013) Typically, as

reported by (Gugliotta et al., 2017) and (Nowacki et al., 2015), SWI is associated with the dry season, and the flood pulse restricts its spread to only a few kilometers during the wet season, compared to tens of kilometers during the dry season This impacts 1.3 million hectares of the VMRD (Carew-Reid, n.d.; Smajgl et al., 2015) In recent research by Van Aalst et al (2023), there is an evident loss in the yield and income of rice farming households in VMRD that links to other socio-economic characteristics under the effect of drought and SWI

In brief, rice production in VMRD is vulnerable to floods, SWI, and climate change due to its geographical features, especially in the coastal areas This is proven to have

an adverse impact on the rice yields of the region

1.2.2.3 Cropping pattern changes and dyke system in Vietnamese Mekong River Delta

In the past, VMRD farmers had to rotate to floating rice crops in flood season, which

is less productive However, with the construction of dikes, farmers can increase their profits with more crops of high-yielding rice in places that were heavily flooded before Therefore, there has been a considerable increase in dyke-protected areas in An Giang This province produces a comparably large amount of rice and other crops in general throughout the years in this region (Nguyen et al., 2019) It can be seen from Figure 1-4 a similar upward trend in the number of dikes constructed and the area of planted upland crops in this province It shows that to increase the rice production and agriculture activity in An Giang, a flood-prone area, according to Wassmann et al (2019), high dykes were constructed widely in the province

in An Giang Province during the period between 2011 and 2014, and the third rice crop covered just over 21,000 ha in 2000, but go up dramatically to 185,000 ha by

2016 (Statistical Office of An Giang Province, 2017) The province of An Giang produced 2.3 million tons of rice in 2000, but by 2013, that number had risen to 4.0 million tons (Statistical Office of An Giang Province 2016) This can mean that a high dike system gives rise to triple-cropping or engender the swift from single/double to triple-cropping pattern

12

Figure 1-5: Dike construction: low and high dikes area in An Giang province

in 2011 and 2014

(Tran et al., 2018) Dyke system plays a vital role in VMRD First of all, it is an adaption to flooding in the region Besides, it allows farmers to practice more productive cropping in flooding areas in VMRD

1.2.3 Sustainable cropping methods in Vietnamese Mekong River Delta

This paper will focus on the cropping methods of shifting between the three main cropping patterns single, double, and triple cropping The research from Chapman and Darby (2016) simulated different cropping methods It concluded that the practice which the farmer performed triple-cropping but shifted to double-cropping and allowed inundation and sediment deposition only in years with higher sediment deposition potential This practice is proven to improve the sediment-bound nutrient, which helps decrease fertilizer use and protect the land from flood while sustaining people’s livelihood

13

Farmers in VMRD are also aware of the detrimental effects of triple-cropping, water, and soil degradation and their impacts on the households’ livelihoods and income (Tran et al., 2018) It is stated that triple-cropping should be discontinued or reduced for environmental benefits, and therefore, the construction of high dikes in the region should also be restrained The focus should shift from quantity to rice quality for exporting purposes

Dike systems are crucial for triple rice cropping An Giang is a very presentative example of this Similar to An Giang, Dong Thap also has a dike system with mainly low dikes and only a few high dikes constructed, unlike An Giang, with mainly high dikes (Tran et al., 2022) In Tong (2017)’s research, where he classified An Giang as intensified cropping with triple crops per year and Dong Thap as balanced cropping with double crops per year, he compared the pros and cons of the two practices (Table 1-1) This shall be discussed further in this paper

Table 1-1: Comparison between costs and benefits of intensified and balanced

cropping in An Giang and Dong Thap, respectively

(Tong, 2017) From the research on the vulnerability and resilience of rice farming households of Van Aalst et al (2023), it is proved that rice farming households do have the capacity

to adapt by diversifying their income, and the least resilient are smallholder and low households, which are unable to do so Therefore, the agricultural transition is one

asset-14

of the practical methods for adapting to VMRD Two main practices for adaptive and beneficial improvement in the region are shifting from rice cropping to rice-aquaculture farming and fruit trees (Bong et al., 2018)

Furthermore, farmers have adjusted their agricultural practices, shifting away from excessive use of fertilizers and pesticides to maximize yield and instead adopting more sustainable production methods that enhance their resilience to climate change Several notable practices have emerged (Tran et al., 2022), such as the "3 Reductions,

3 Gains" approach, the "1 Must Do, 5 Reductions" approach, and the "1 Must Do, 6 Reductions" approach explained in Table 1-2 Moreover, there has been an increasing adoption of economically efficient irrigation techniques, as well as the implementation

of VietGAP and GlobalGAP standards in farming operations (Bong et al., 2018) Table 1-2: Explanation for sustainable cropping practice approaches in VMRD

3 reductions, 3 gains Reductions: seed rate, pesticides, and Nitrogen fertilizers

used; Gains: Improved productivity, quality, and economic efficiency/income

1 must do, 5 reductions Must do: using certified seed; Reductions: seed rate,

pesticides, fertilizers, water used, and harvest loss

1 must do, 6 reductions Must Do: using certified seed; Reductions: seed rate sowing,

pesticides, fertilizers, water used, harvest loss, and greenhouse emissions

producing methods for clean and safe products, especially fresh fruits and vegetables Four main areas are listed for regulations: the producing technologies, the standard for food safety and hygiene, the standard for the labor and

workplace, and product sources and origins (GOOD AGRICULTURE PRACTICE - VietGAP Standard, 2023)

15

stricter regulations with 252 criteria, of which 36 must be followed 100%, 127 ones must be followed at 95%, and 89 ones are recommended Products with these standards are accepted worldwide and can be traded in any kind of market (VietGAP Và GlobalGAP Là Gì? Lợi Ích Từ Quy Trình GAP, n.d.)

1.3 Scope of the research

This research will analyze the remote sensing data for the VMRD region, which consists of 13 provincial units (including Can Tho city), from 2000 to 2022 This data will help enlighten information about the cropping patterns as well as the climate-risk effects on the region

Narrowing down multiple climate risks into two main ones: flooding and SWI, this paper will analyze and discuss the two provinces that typically suffer from the risks:

An Giang (flooding) and Tra Vinh (SWI) Through the collected data, an assessment

of whether the cropping patterns or the cropping pattern changes are successful or not will be made by consulting from formerly relevant reports and research

Since rice is the dominant crop in VMRD, this paper will also focus on this crop to analyze the rice cropping pattern adaptability to climate risks in the region

Thereby, the research will mainly discuss the effects and future recommendations for VMRD areas' rice cropping production

1.4 Research questions and hypotheses

The questions and hypotheses of this research are summarised in the table below:

Table 1-3:Researching questions and hypotheses

16

How do the cropping

patterns change

throughout the years?

From the remote sensing data, we can see the cropping pattern change throughout the years We shall see the increase/ decrease of the triple-cropping pattern in different areas

How do the cropping

patterns changes relate to

the climate risks?

The cropping patterns changes are different from different risk-prone areas and also result from the LULC changes where there are dikes or where there is land erosion/SWI

What are the optimal

measures for adaptation in

the cropping pattern to

climate risks in VMRD?

The triple-cropping pattern is still debatable for its benefits Therefore, it is essential to discuss the pros and cons of this and recommend optimal practices for rice cropping in VMRD

1.5 Research objectives

From the research questions, hypotheses above, and the scope, this research targets the following objectives:

 Objective 1: Analyse the cropping pattern from time-series remote sensing data

o Collect and process data

o Use software to visualize and extract data

o Analyze the data

o Analyze the changes in cropping systems

 Objective 2: Analyse the climate-risks in VMRD

o Collect data about climate-risks in VMRD: flooding and SWI from remote sensing data and previous research assessment

o Categorize the areas into different disaster-prone areas

o Compare and analyze the differences between An Giang (flood-prone) and Tra Vinh (SWI-prone)

 Objective 3: Assessing the impact of climate risks on the cropping patterns in

An Giang and Tra Vinh

 Objective 4: Propose resilient cropping practices

o Analyze the efficiency and problems of triple cropping pattern in VMRD

o Recommend some approach for more sustainable cropping practices in VMRD

1.6 Study area

The VMRD, located in southern VN, boasts a coastline exceeding 732km, making it the largest delta in the region It is surrounded by the Gulf of Thailand in the west and the East Sea in the east and borders Cambodia to the north, with Ho Chi Minh City situated in the northwest The delta includes one city, Can Tho, and 12 provinces, namely Long An, Tien Giang, Ben Tre, Vinh Long, Tra Vinh, Hau Giang, Soc Trang, Dong Thap, An Giang, Kien Giang, Bạc Lieu, and Ca Mau

18

Figure 1-6: VMRD administrative map

As per the Vietnamese General Statistics Office’s yearbook of 2020, the population of VMRD, which covers 12.32% of Vietnam's total terrestrial area, is 17,318,600 people with a total area of 40,816 km2 Agriculture land in VMRD covers 2.615 million ha and represents 64% of the total natural area, with paddy cultivation and aquaculture as the main activities Additionally, the agricultural sector in VMRD employed 43.3% of the country's overall population in 2020 The terrain is mostly flat, sloping downward from North to South, and includes two vast low-lying regions: Dong Thap Muoi and the Northeast of the Ca Mau peninsula As a result, VMRD has rich plains, a dense canal system, and a lengthy coastline, making it an ideal location for agriculture VMRD is renowned as "Vietnam's Rice Bowl," producing over 40% of the country's agricultural productivity, 54% of rice production, 90% of export rice production, as well as numerous aquatic and fruit goods In 2019, VMRD contributed 17.7% to the country's GDP, according to Vu et al (2020)

The natural environment, especially the climate of the VMRD, can be summarized below (SCIENCE & DEVELOPMENT (VESDEC), n.d.) The temperature in the Mekong Delta region is consistent and high throughout the year, ranging from 25 - 27°C, with an average of 6-7 hours of sunshine per day and an average evaporation

19

rate of 1,000 - 1,300mm The air has a relative humidity of 78-82% and an average annual temperature of 27.5 - 27.8°C March and April are the hottest months of the year, with daytime temperatures ranging from 28 - 35°C From June to February, the temperature is lower, but usually not below 20°C

From 2000 to 2020, the Mekong Delta received an average of 1,733mm of rainfall annually, with heavy rainfall concentrated in the Ca Mau peninsula, ranging from 1,400 to 2,200mm The region's total amount of rainwater resources is 65.4 billion m3/year In the last 20 years, some stations have experienced a 10-15% decrease in annual rainfall

Rainfall in the region varies spatially, with the highest average annual rainfall occurring on the West Coast, ranging from 2,000-2,400mm in Camau and South Kiengiang, and gradually decreasing on the East Coast, from 1,700-1,800mm in Bac Lieu and Soc Trang The area between the Hau and Tien rivers, such as Angiang and Tiengiang, has the lowest average annual rainfall, while the Northeast has an average

of 1,600-1,800mm in Tan An and Moc Hoa

For the agriculture development plan for the VMRD, the region aims to decrease rice production, increase fruit yield and aquaculture, and develop ecotourism The direction for the development plan in VMRD has been approved by the government to adjust the organization of agricultural production in order to accommodate changes in environmental conditions in accordance with the three ecological sub-regions as below:

20

Figure 1-7: 3 ecological sub-regions for agricultural development plan in VMRD

(adapted from (Mekong Delta Plan, 2013))

Table 1-4: Approved agriculture development plan in VMRD period of 2011-2025

Ecological zone Administrative areas Strategic agriculture

development plan

Freshwater ecological

zone in the upstream

and central part of the

delta

Vinh Long, Can Tho city and a section of

Trang, Tra Vinh, Ben Tre, Tien Giang, Long

An

Protected from the impact of floods, inundation, and SWI, it is

a crucial area for the production

of rice, freshwater aquatic products, and fruits of the Mekong Delta and the country; o create a varied, innovative, and sustainable agriculture that takes into account adaptation to severe

Developing marine navigation Specialized brackish water production and rotation with rice and vegetables suitable to seasonal water conditions

brackish ecological

zones

Part of Kien Giang,

Ca Mau, Bac Lieu, Soc Trang, Tra Vinh, Ben Tre, Tien Giang, and Long An

saltwater, saline-brackish water products on the shore and in the sea; fishing; restoring and developing coastal mangrove forests in conjunction with biodiversity protection and coastal strip; develop the agro-forestry system in the direction

of ecology, organic, combined with ecotourism; proactively prevent, avoid and minimize the risks of natural disasters, climate change, and sea level rise

(adapted from (Chi, 2022))

As from Figure 1-2 from Wassmann et al (2019), the two focused areas of this research suffer from different climate risks as An Giang is mainly affected by flooding and Tra Vinh by SWI Tra Vinh is located on the low-lying coast of VN and is one of the most vulnerable provinces to SWI in VMRD and in VN, which affect badly to the

22

province's agricultural production (Dang et al., 2020; Karila et al., 2014; Nguyen et al., 2020) The impacts of SWI are devastating as it creates a hostile environment for the metabolism of soil organisms leading to soil degradation, thus damaging all the plantations and other organisms in the soil According to (Nguyen et al., 2020), it is unavoidable that CC impacts such as sea level rise, SWI, and drought all have an adverse impact on Tra Vinh province, leading to the increase of soil salinity in the area, particularly in lower-lying terrains such as estuaries and coastal areas (Figure 1-8) As a result, Tra Vinh Province's agricultural cultivation has changed For Tra Vinh Province's agriculture and economic development to continue, measures to adapt

to increased soil salinity need development

Figure 1-8: SWI leading to soil salinity map with the depth from 0 cm to 20 cm

in Tra Vinh province, VMRD, VN (Nguyen et al., 2020)

An Giang, with a different location and condition, suffers from different climate risks from Tra Vinh (Figure 1-2) This province belongs to the upstream part of the Lower Mekong River basin in VN, and flooding is a common phenomenon here but is controlled by the dike infrastructure system, as shown in Figure 1-10 (Nguyen et al., 2019) Thanks to this system, farmers in this province are able to practice intensified cropping, which produces more than 80% rice in the whole delta (Chapman & Darby, 2016)

23

Figure 1-9: An Giang and Tra Vinh in administrative map

Figure 1-10: Dike area in An Giang Province, VN, in 2014

(Nguyen et al., 2019)

1.7 The framework of the research

This research is conducted based on the logical framework in Figure 1-11 below The data is collected, processed, and analyzed remotely but validated by the field-trip The research study examines the correlation between changes in cropping patterns and climate risks It assesses the adaptability of cropping patterns in mitigating these risks and identifies suitable practices for different climate-affected areas By investigating

24

these three aspects, the study provides valuable insights into the dynamics and

effectiveness of cropping pattern adaptations in response to climate risks

Figure 1-11: The logical framework of the research

25

CHAPTER 2 DATA AND METHODS

2.1 Data used

Listed below are the data used in this research:

- Topographical data: The river network and LULC change maps are collected from the regional and provincial statistical reports

- Climate risks data: From desk review method and remote sensing data

- Remote sensing data: From remote sensing data, the 8-day-time-series data collected from MODIS (Moderate Resolution Imaging Spectroradiometer) and LANDSAT observation satellite systems, we can collect EVI and LSWI data to analyze the crop frequency, inundation period, sowing-heading-harvesting dates Besides, LULC data are collected from LANDSAT to analyze the LULC changes throughout 2001-2021

- Social development data: Similar to topographical data, from the regional and provincial statistical reports, we can collect data about the population, population distribution, economic development, and social development in the area In this paper, the analysis of crop production is emphasized Additionally, in-depth interviews with the people are also implemented to understand the cropping pattern changes, their response, and their recommendation for adapting to CC

2.2.1 Remote sensing data collection and analysis

Remote sensing data is collected from LANDSAT to reflect the changes in LULC in the VMRD from 2001 to 2021 From the satellite images of the study area, each pixel

is classified into its categories based on similar spectral characteristics This study uses the classification scheme of the IGBP (International Geosphere-Biosphere

26

Programme) to classify LULC It categorizes land cover into 17 classes with the color code presented below

Table 2-1: IGBP classification scheme and RGB color code for LULC map

Name Value RGB code

Evergreen Needleleaf Forests 1 rgb(2, 100, 1)

Evergreen Broadleaf Forests 2 rgb(1, 130, 1)

Deciduous Needleleaf Forests 3 rgb(151, 190, 71)

Deciduous Broadleaf Forests 4 rgb(2, 221, 2)

Urban and Built-up Lands 13 rgb(0, 255, 255)

Cropland/Natural Vegetation Mo-

27

Other than that, time-series satellite image analysis technology is used to evaluate the changes in cropping patterns and flooding situations The satellite image data used in this research are observation data obtained by MODIS (Moderate Resolution Imaging Spectroradiometer) MODIS is the name of the optical sensor mounted on NASA's earth observation satellites EOS AM-1 (commonly known as Terra) and PM-1 (commonly known as Aqua) Both satellites were launched in December 1999 for Terra and in May 2002 for Aqua Observations are still ongoing as of 2022 The Terra and Aqua satellites cross the equator at 10:30 AM and 1:30 PM in two different nodes, providing two images of the study area daily (Kotera et al., 2015) MODIS observes the range of 0.4 to 14 μm with 36 bands It is widely used to observe snow cover, temperature, humidity, sea ice, etc Spatial resolutions are 250 m (bands 1 and 2), 500

m (bands 3-7), and 1000 m (bands 8-36) Although the number of return days is 16, the observation swath is relatively wide at 2,330 km (±55 degrees), so it is possible to acquire images of the same point almost daily In this study, the pixel resolution is 250m*250, and the MODIS noiseless data product MCD19Q2 was used, including 250m-8 days of NDVI, EVI, and LSWI (Kotera et al., 2015) Presented in Table 2-2 is the date of the year (DOY) table based on the MODIS data acquisition date of some years for demonstration The detailed acquisition date table is attached in Appendix F The image is taken every eight days throughout the whole period

Table 2-2: MODIS data acquisition date table

29

as a value obtained by normalizing the difference in reflectance between the red region and the near-infrared region The vegetation situation can be captured in the study area this way

In this study, Enhanced Vegetation Index is used as a vegetation index (EVI) was used EVI is calculated by Equation (1) using MODIS's ground surface reflection data

(1) The spectral spotting of EVI helps indicates the cropping frequency in a year The X-axis indicates the days throughout the whole observed period Looking at Figure 2-1, each peak in the EVI spectral viewer represents one crop as the area’s EVI reaches its peaks only once during the one crop Therefore, counting the number of peaks can give us an estimation of the number of crops practiced in the year

Figure 2-1: EVI spectral viewer Here, NIR is the reflectance of the near-infrared region, RED is the red region of the visible light region, and BLUE is the blue wavelength region Blue wavelengths are used in EVI to correct atmospheric effects Parameters C1 and C2 are correction factors for aerosols in the atmosphere, L is the correction factor for the vegetation canopy background, and G is the gain factor, where L=1, C1=6, C2=7.5, and G=2.5, respectively

Besides, to detect inundation periods or flooding situations, we need the combination

of EVI and the Land Surface Water Index (LSWI) High sensitivity to moisture content is related to the surface reflectance value in the short-wave infrared band – band 6 in the LSWI calculation (Sakamoto et al., 2009)

2.2.2 Ground truth observation

The ground truth observation (GTO) method is a process of validating or verifying the accuracy of remotely sensed data by collecting actual measurements and observations

on the ground This method involves physically visiting the site or area being studied and collecting data on land cover, vegetation, topography, and other features of interest The data collected from GTO can be used to calibrate and validate remote sensing data and develop accurate land cover maps GTO is particularly important in areas with complex terrain or mixed land use, where remote sensing data may be difficult to interpret or prone to errors Using the GTO method can improve the accuracy and validity of data from remote sensing for a range of uses, such as environmental monitoring, land use planning, and natural resource management

In this study, the author visited VMRD for onsite observation with the schedule and route described in Table 1-1 and Figure 2-2 below During the field trip, a GoPro camera was used to capture the scene from the side of the road, and GPS plugins from QGIS were also connected to the camera to record the coordinates of the media (photos, videos) accordingly With this primitive equipment set, only media from the roadside are collected to compare with the remote sensing data As the route was fixed because the author can only observe on and inside the vehicle, not all LULC types or cropping pattern areas were visited Some points were selected randomly to compare with the land use and landcover and cropping pattern map created from remote sensing data to validate the results from remote sensing data

31

Table 2-3: GTO field trip schedule in VMRD (March 2023)

Date 22/03/2023 23/03/2023 24/03/2023 25/03/2023

Morning Fieldwork in Tra

Vinh

Fieldwork in Soc Trang

Fieldwork in An Giang

Fieldwork from Long An to HCMC

Afternoon Fieldwork in Tra

Vinh

Fieldwork in Soc Trang

Fieldwork in An Giang

Fieldwork from Long An to HCMC

Evening Go back to Can Tho Go back to Can Tho Go back to Can Tho Go back to Ha Noi

Figure 2-2: GTO in VMRD route (blue line) For validating the remote sensing data, pictures of places were taken with GPS coordinates and then compared with the raster data of the coordinates accordingly in QGIS Figure 2-3 illustrates the process of comparing for validating First, with the photo taken during the GTO field trip, we compare it with the land use data and other indexes to see whether the remote sensing data is accurate in reality