Vulnerability to Flood in the Vietnamese Mekong ...

Journal of Environmental Science and Engineering B 2 (2013) 229-237 Formerly part of Journal of Environmental Science and Engineering, ISSN 1934-8932

D

DAVID

PUBLISHING

Vulnerability to Flood in the Vietnamese Mekong Delta: Mapping and Uncertainty Assessment Van Pham Dang Tri, Nguyen Hieu Trung and Vo Quoc Thanh College of Environment and Natural Resources, Can Tho University, Can Tho City, Vietnam Received: December 25, 2012 / Accepted: January15, 2013 / Published: April 20, 2013. Abstract: The Vietnamese Mekong Delta is located at the end of the Mekong River, one of the 10 largest rivers in the world. It plays an important role, especially in terms of food security for not only Vietnam but also the world. However, the Vietnamese Mekong Delta is projected to be heavily affected by: (1) the annual (fluvial) flood, which would be changed in terms of time and spatial distribution after impacts of climate change scenarios (i.e., sharper hydrograph with shorter flood period); and (2) sea level rise. Such combination would result in significant changes of surface water resources, leading to consequent impacts on the existing farming systems in the Vietnamese Mekong Delta. Therefore, this paper presents a new approach of integrating a one-dimensional hydrodynamic model (ISIS-1D) with GIS (Geographic Information System ) analyses to: (1) identify priority areas for flood adaptation and mitigation; (2) provide an insight to local decision-makers in the Vietnamese Mekong Delta in changes of future floods. Keywords: Vulnerability, uncertainty, flood, Vietnamese Mekong delta.

1. Introduction The Vietnamese Mekong Delta is located at the end of the Mekong River (Fig. 1), one of the 10 largest rivers in the world and running through six countries (China, Myanmar, Thailand, Laos, Cambodia and Vietnam) [1]. The Vietnamese Mekong Delta provides annually 90% of national rice export and contributes significant aquaculture production for the national and global food market [2]. According to Thien [3], the main causes of floods in the delta include: (1) flood discharge from the upstream; (2) local heavy rainfall (driven by monsoon or typhoons); (3) high tides in the East Sea and West Sea. The annual fluvial floods (the phenomenon considered in this paper) are a common natural event and classified into two types, including: (1) floods caused by the upstream discharge with long flood inundation from 2 to 6 months (ranging from July to December [4]); (2) tidal-induced flood driven 

Corresponding author: Van Pham Dang Tri, Ph.D., research fields: hydrodynamic modelling, waterscape management and (surface) water resource changes. E-mail: [email protected].

by tidal regimes in the East Sea and West Sea [3]. During the flood season, water enters the Vietnamese Mekong Delta via the main river reaches (the Mekong—running through Tan Chau and My Thuan to the East Sea and Bassac—running through Chau Doc and Can Tho to the East Sea) (Fig. 1) and overflows across the common border between Vietnam and Cambodia [5, 6]. The annual fluvial flood brings significant benefits to the delta. During the flood season, the annual flood conveys about 160 Mt of sediment per year [7] and a large amount fish (in average, about 475.73 t) with about 1,200 fish species in total to the delta [8]. In specific, the 2000-flood (the 20-yr returned-period event [9]) brought about 1.86 million tons of fish (approximately 2,600 million USD) [10]. Besides, annual flood plays an important role for wetland protection and biodiversity conservation [4]. However, the annual flood also causes negative impacts on the livelihood of local residents (e.g., losses of life and properties). For example, the 2000-flood caused approximately 250 million USD of total damages

230

Fig. 1

Vulnerability to Flood in the Vietnamese Mekong Delta: Mapping and Uncertainty Assessment

The Vietnamese Mekong Delta [6] and its provinces.

[10]. With great impacts of the global climate change, the local hydrological conditions in the Mekong Basin in general and the delta in specific are projected to be significantly changed [6], leading to a requirement of vulnerability assessment to define priority areas for flood mitigation and adaptation. This

paper

is

to

characterize

the

hazard,

vulnerability and risk caused by the extreme historical flood in the Vietnamese Mekong Delta (in 2000) and the projected ones in 2050 [6] with changes of upstream discharge and sea level rise.

2. Materials and Methods 2.1. Hydrodynamic Model Setup In this study, the available ISIS-1D model (developed by the Mekong River Commission) for the entire Mekong Delta (including the Cambodia and Vietnamese parts) is used with application of the hydrodynamic component but not the hydrological one. In fact, this model is used by the MRC (Mekong River Commission ) to determine the annual fluvial flood in the Mekong Delta. The hydrographs at the upstream boundary (in Kratie, Cambodia) are

determined based on interpolated historical discharge and predicted ones as per the climate change scenarios CC1 and CC2 (Fig. 2). According to the CC1 climate change projection, the Mekong River discharge in Kratie is greater than the one of the CC2 projection; in fact, with future developments (hydropower and irrigation) in the upstream Mekong basin, less water would arrive in the delta [6]. In addition, the sea level in the future is the projected rise (+30 cm for both the East Sea and West Sea with reference to that in 2000 [11]). Table 1 summarizes scenarios applied for the hydrodynamic model in this paper. Even though the ISIS-1D model is considered to be able to provide a reasonable representation of the hydrodynamics of the Cambodian floodplain and Vietnamese Mekong Delta, MRC suggests that the model should not be used for design purposes [12], but rather to estimate the trend of changes when the boundary conditions are modified. The reason for such advise is that, for the large (deltaic) scale model, the lack of details for a specific area will lead to underand/or over-estimation of the future events, and consequently wrong calculations of the planned construction.


231

Fig. 2 Measured hourly discharge in 2000 at Tan Chau and Chau Doc: (A) Annual hydrograph measured in 2000, historical mean daily discharge (1985-2000) and projected annual hydrograph in 2050 (Scenarios 1 and 2); (B) Measured and projected daily discharge in Kratie, Cambodia [6]. Table 1

Setup of the model boundary conditions.

Models setup

Upstream boundary conditions

Downstream boundary conditions Sea water level in year 2000 (SL 2000) for both the West Sea and East Base-line scenario Discharge hydrograph of the year 2000 Sea Scenario 1a Projected discharge of the year 2050 (CC1) Sea water level in year 2050 (SL 2000 + sea level rise) Scenario 1b

Projected discharge of the year 2050 (CC2) Sea water level in year 2050 (SL 2000 + sea level rise)

2.2 Hazard Mapping Flood hazard is categorized based on the level of difficulties in daily life and/or damage of properties [13]. The flood hazard maps are created according to the flood depths (in comparison to the local land surface elevation) (Table 2). According to the previous study (e.g., Refs. [6, 14]), the fluvial flood in the Vietnamese Mekong Delta should be characterized with great attention paid to the upstream discharge driven flood and tidal-induced flood; therefore, two specific days are selected to create inundated maps, including: (1) the highest flood stage

in the upstream section according to the greatest measured stages at Tan Chau and Chau Doc; (2) the largest flood extent at the deltaic scale based on the greatest measured stage corresponding to the greatest stages measured at Can Tho and My Thuan. Table 2

Classification of flood hazards.

Water depth (m) 0-0.2 0.2-0.5 0.5-1.0 1.0-2.0 > 2.0

Hazard ranking 0-0.04 0.04-0.1 0.1-0.2 0.2-0.4 0.4-1

Hazard category Very low Low Medium High Very high

232


2.3 Vulnerabilities and Risks Assessment The Coastal City Flood Vulnerability Index (based on exposure, susceptibility and resilience to coastal flooding indicators) [15, 16] are modified and applied to meet the actual conditions of the Vietnamese Mekong Delta and each province in the delta is identified as a calculated unit. The considered components to evaluate the vulnerabilities and risks of each calculated unit in the delta include the hydro-geological and climatic components (i.e., sea-level rise, river discharge, soil subsidence, number of cyclones, storm surge), socio-economic components (a part of the socio-economic system) and politico-administrative components (the administrative and institutional system) [16]. With limitation of the available data of socio-economic for the future, the vulnerability of different provinces in the delta is calculated for the current socio-economic conditions in conjunction with projection of upstream discharge in 2050 and sea level rise of 30 cm (with the reference of daily measured sea level in 2000). The calculated indicators are standardized after Eqs. (1) and (2) for the positive and negative impacts, respectively. The standardized indicators are multiplied with weights (ranging from 1 to 10) of each indicator based on their importance to vulnerability in specific conditions of the Vietnamese Mekong Delta

Fig. 3

(according to the available publications and local reports, e.g., Refs. [17, 18]). All indicators are then aggregated to identify vulnerability of each calculated unit Eq. (3). Vulnerability uncertainty assessment is done by adjusting the applied weight according to Eq. (4) and the following rules: (1) For each iteration, weight of one indicator would be adjusted while the others would be remained; (2) The iteration would keep going until all assigned weights are changed. According to Ref. [19], risk value is the product of vulnerability and hazard with 0 and 1 corresponding to the lowest and greatest. x 1 x Exposure Susceptibility Resilience New_Weight Assigned_Weight 1

x Vulnerability

(1) (2) (3) (4)

3. Results 3.1 Simulated Flood Fig. 3 presents the simulated stages of the fluvial flood at the Tan Chau, Chau Doc, My Thuan and Can Tho gauging stations from September 15 to October 15, 2000. The greatest (simulated vs. measured) stages in Tan Chau (5.18 m vs. 5.06 m above mean sea level) and Chau Doc (4.73 m vs. 4.90 m above mean sea level) appeared on September 23, 2000 (Fig. 4a) while such the greatest in Can Tho (2.31 m vs. 1.79 m above mean sea level) and My Thuan (2.23 m vs. 1.80 m

Simulated stages at Can Tho, My Thuan, Chau Doc and Tan Chau; msl is mean sea level.


233

Fig. 4

Inundated map at the peak flood (a) and greatest inundated area (b) in the Vietnamese Mekong Delta.

Fig. 5

Hazard maps at the greatest stage in the upstream (a) and greatest flood extent in the Vietnamese Mekong Delta (b).

above mean sea level) appeared on September 28, 2000 (Fig. 4b) [6, 10, 20]. Even though there are differences between the simulated and measured stages, the differences are among the accepted ranges [6]. 3.2 Hazard Map The hazard maps of flood in 2000 are shown in Fig. 5. Despite the greatest stages appeared in Tan Chau and Chau Doc, influenced area is not the greatest (Fig. 5a) as the flood is still remained in the upstream section of the delta. Flood hazard covers about 48% of the entire Vietnamese Mekong Delta, including very low, low, medium, high, and very high hazard area

corresponding to 7.89%, 5.71%, 8.77%, 12.93% and 12.71% of total area, respectively. The high and very high hazards are mainly found in the upstream, while there is improbable hazard in the coastal area. However, when the flood is expanded largest in the delta (Fig. 5b), the medium, high, and very high hazard areas do not only expand in the upstream but also in the coastal areas. The influenced area are 5.98%, 12.67%, 17.84%, 25.84% and 14.29% of the entire delta, accounted for very low, low, medium, high, and very high hazard, respectively. In 2050, simulated greatest flood stage would appear in the upstream section of the delta on October 16 and the flood extent would be the largest in the

234


Fig. 6 Hazard maps in Scenario 1 at the greatest stage in the upstream and greatest flood extent in the Vietnamese Mekong Delta.

following days. As the flood dynamics are not significantly different between the two scenarios (of the projected upstream discharge), Fig. 6 presents hazard maps of the greatest flood stage in the upstream section of the delta and at the greatest extend according to Scenario 1 only. At the greatest flood stage in the upstream section, the hazard area would cover 60.38 % of the entire delta, including the very low, low, medium, high and very high hazard

lower flood peaks in the future scenarios. At the greatest flood extent in the future, the hazard area would cover 81.45 % and 79.21 % of the entire delta in the Scenario 1 and Scenario 2, respectively. The high flood hazard does not only concentrate in the upstream but also open to the downstream accounting for about 30% area of the entire delta. 3.3 Vulnerabilities

corresponding to 7.52%, 15.31%, 16.15%, 16.83%

Fig. 7a introduces the vulnerability of each province

and 4.58% of the entire area, respectively. The high

in the Vietnamese Mekong Delta in the base-line

and very high hazard would be found also in the

scenario with the calculated integrated Coastal City

upstream section of the delta. Similarly, in Scenario

Flood Vulnerability Index. In general, differences in

2, the greatest flood stage in the upstream section of

flood-induced vulnerability between provinces are

the delta and the greatest flood extent would also

relatively small as the provinces share a common

appear on the same day to Scenario 1. The hazard

system

areas in Scenario 2 (59.68%) would be smaller than

unemployment level and population density). The Soc

those in Scenario 1 as the upstream discharge in

Trang Province would be most vulnerable area as the

Scenario 2 is projected to be smaller than that in

area is facing directly to the Bassac River and the East

Scenario 1. In comparison to the 2000-flood, the flood hazard in the future would open further to the Ca Mau Peninsula as the consequence of the projected sea level rise. The simulated hazard areas would increase about 12.39% and 11.68% in Scenario 1 and 2, respectively. Although hazard areas would increase, very high hazard area would decrease according to the

Sea, having a dense of the canal network and less ratio

of

the

delta

(e.g.,

cultural

heritage,

of investment over the total gross domestic product. The Bac Lieu Province is the second most vulnerable to flood because of high population growing, a dense canal network and facing directly to the East Sea. The following vulnerable province is accounted for Can Tho and Hau Giang with a dense river network and less ratio of investment over the total gross domestic


235

Fig. 7 Vulnerability of the Vietnamese Mekong Delta (a) and range of flood vulnerability (by province) after changes of the assigned weight (b).

Fig. 8

Risk maps of Scenario 1 at the peak flood (a) and greatest flood extent (b) in the Vietnamese Mekong Delta.

product; in addition, Can Tho is facing directly to the Bassac River. Other coastal provinces (Ca Mau, Ben Tre, Tra Vinh and Tien Giang) would be directly affected by sea level rise and the increase of storm surge in the future but with less magnitude in comparison to the other provinces. In addition, local residents of those provinces receive little impacts (if any) from the annual flood, so they would not have sufficient experiences for flood adaptation. The low vulnerable province is Ben Tre with the low population density considered as the most important indicator. An Giang and Dong Thap are the two least vulnerable provinces as they are not facing directly to the sea and therefore the sea level (rise) does not introduce strong influences on the area [6, 14, 21]. In addition, local residents of the two provinces have

sufficient knowledge and experiences on living-with-flood. Fig. 7b illustrates the range of vulnerability by province after the changes of the assigned weight. Soc Trang, Bac Lieu, Tien Giang, Tra Vinh and Ben Tre are considered to be highly vulnerable, while An Giang and Dong Thap are of the low vulnerability values. Although there are the changes of flood vulnerability, Soc Trang is also of the greatest vulnerable area according to the annual fluvial flood. As the flood dynamics generated for Scenarios 1 and 2 are not significantly different, Fig. 8 only presents the risk map of Scenario 1 at the peak and the greatest flood extent. At the greatest flood stage in Tan Chau and Chau Doc, the percentages of flood risk are of about 60% and 59% area of the entire delta in

236


Scenarios 1 and 2, respectively. When the flood expands largest spatially, the flood risk would cover most of the entire area of the entire delta. In addition, the high category of flood risk would concentrate along the main water-ways (including the Mekong and Bassac River).

4. Conclusions With the application of the deltaic-scale model, differences between the measured and simulated stages are still large. In addition, the hydraulic constructions system and river bathymetry are not updated fully leading to a need to update the model, especially (1) the full-dyke system in the upstream section (to prevent flooding the intensive rice field); (2) sluices along the coastline (to prevent saline intrusion and also to confine the flood discharge loaded to the sea); (3) current status of the river network and actual status of bathymetry. The characterization of the historical flood hazard, vulnerability and risk are quantified. Even though the flood depth and duration in the upstream section of the Vietnamese Mekong Delta are greater than those in the downstream section, the risk of local residents due to the annual flood in the upstream is lower than that in the downstream. In fact, local residents living in the upstream section have great experiences to eliminate negative impacts and earn the most benefits from such natural events. In contrast, local residents living in downstream section of the delta are not used to tidal-induced floods leading to negative impacts (e.g., damages of the agriculture and aquaculture [22], leading to negative impacts on the local livelihood [23, 24]. The hazard in 2050 would increase, especially in the downstream section of the delta due to the projected sea level rise. The results of this study will help local stakeholders and decision-makers understand what might happen in the future in terms of hydrological changes and consequent impacts (i.e., risk) on each province. The obtained results also

introduce local areas that need specific attention for a detailed study in the future. The coastal areas, especially those along the East Sea, are projected to be highly affected due to the hydrological changes while there is still limitation of awareness for flood adaptation. Therefore, future research is required to provide suitable set of indicators that meet the actual conditions of the Vietnamese Mekong Delta and detailed study should be done to meet specific local settings. Finally, the study paid great attention to the magnitude of flood and its impacts, however, does not take into account the timing distribution of flood dynamics. Such dynamics, which may have great impacts on the agriculture and aquaculture activities of the local residents, should be studied in more details in the near future.

Acknowledgments This paper is developed with financial support from the Ministry of Economic Affairs from the Netherlands with the project of Developing Agriculture, Aquaculture and Environment based Climate Change Adaptation strategies for the Mekong Delta Plan of Vietnam. This project was carried out with support of experts from Wageningen University and Research, the Netherlands and Can Tho University, Vietnam.

References [1] [2]

[3]

[4] [5]

State of the Basin Report 2010, MRC (Mekong River Commission), Vientiane, Laos, 2010. Statistical Yearbook of Vietnam 2009, VGSO (Vietnamese General Statistics Office), Statistical Publishing House of Vietnam, 2010. L.T.T. Thien, Review of flooding and flood management: Viet Nam country statement, Technical session I: Introduction to flood management, in: Flood Management and Mitigation in the Mekong River Basin, FAO, RAP Publication, 1999. Annual Mekong Flood Report 2008, The MRC (Mekong River Commission), Vientiane, Laos, 2009. Overview of the Hydrology of the Mekong Basin, The MRC (Mekong River Commission), Vientiane, Laos,


[6]

[7] [8]

[9]

[10]

[11]

[12]

[13]

[14]

2005. V.P.D. Tri, I. Popescu, A. van-Griensven, D. Solomatine, N.H. Trung, A. Green, A study of the climate change impacts on fluvial flood propagation in the Vietnamese Mekong Delta, Hydrol., Earth Syst. Sci. Discuss. 9 (2013) 7227-7270. D.E. Walling, The changing sediment load of the Mekong River, Ambio 37 (2008) 150-157. D. Blackmor, C. Perry, R. Stein, Modelled Observations on Development Scenarios in the Lower Mekong Basin, A tech nical report for the World Bank’s Mekong Regional Water Resources Assistance Strategy, World Bank, 2004. L.T.V. Hoa, N.H. Nhan, E. Wolanski, T.T. Cong, S. Haruyama, The combined impact on the flooding in Vietnam's Mekong River delta of local man-made structures, sea level rise and dams upstream in the river catchment, Estuarine, Coastal and Shelf Science 71 (2007) 110-116. L.A. Tuấn, C.T. Hoành, F. Miller, B.T. Sinh, Flood and salinity management in the Mekong Delta, Vietnam, in: T.T. Be, BachTan Sinh, F. Miller (Eds.), Challenges to Sustainable Development in the Mekong Delta: Regional and National Policy Issues and Research Needs, The Sustainable Mekong Research Network, 2007. Scenarios for Climate Change and Sea Level Rise in Vietna, MONRE (Ministry of Natural Resources and Environment), Ha Noi, Viet Nam, 2011. Improved Flood Risk Reduction through Implementation of Structural Measures Flood Proofing, MRC (Mekong River Commission), Vientiane, Laos, 2010. T. Tingsanchali, M.F. Karim, Flood hazard and risk analysis in the southwest region of Bangladesh, Hydrological Processes 19 (2005) 2055-2069. R. Wassmann, N.X. Hien, C.T. Hoanh, T.P. Tuong, Sea level rise affecting the vietnamese Mekong Delta: Water elevation in the flood season and implications for rice

237

production, Climatic Change 66 (2004) 89-107. [15] S. Balica, N.G. Wright, Reducing the complexity of the flood vulnerability index, Environmental Hazards 9 (2010) 321-339. [16] S.F. Balica, N.G. Wright, F. Meulen, A flood vulnerability index for coastal cities and its use in assessing climate change impacts, Natural Hazards 64 (2012) 73-105. [17] Q. Dinh, S. Balica, I. Popescu, A. Jonoski, Climate change impact on flood hazard, vulnerability and risk of the Long Xuyen Quadrangle in the Mekong Delta, International Journal of River Basin Management 10 (2012) 103-120. [18] Socio-economic Conditions of the Vietnamese Mekong Delta from 2001 to 2011, Can Tho University Publishing House, Can Tho, Vietnam, 2012. [19] Mitigating natural disasters, phenomena, effects and options, A Manual for Policy Makers United Nations, New York, USA, 1991. [20] H.T.L. Huong, A. Pathirana, Urbanization and climate change impacts on future urban flooding in Can Tho city, Vietnam, Hydrology and Earth System Sciences 17 (2013) 379-394. [21] K. Västilä, M. Kummu, C. Sangmanee, S. Chinvanno, Modelling climate change impacts on the flood pulse in the Lower Mekong floodplains, Journal of Water and Climate Change 1 (2010) 67. [22] N.H. Ninh, V.K. Trung, N.X. Niem, Flooding in Mekong River Delta, Vietnam, in: Human Development Report 2007/2008 Fighting Climate Change: Human Solidarity In A Divided World, UNDP, 2007. [23] M. Kakonen, Mekong delta at the crossroads, more control or adaptation? Ambio 37 (2008) 205-212. [24] J. Manuta, L. Lebel, Climate Change and the Risks of Flood Disasters in Asia: Crafting Adaptive and Just Institutions, in: An International workshop: Human Security and Climate Change, Asker, Norway, 2005.