Mike Flood Application for Solving Inundation Issues for Ho Chi Minh City in The Context of Climate Change: A Case Study in the District 145221

Mike Flood Application for Solving Inundation Issues for Ho Chi Minh City in The Context of Climate Change: A Case Study in the District 1 Pham Thanh Long 1* , Tran Tuan Hoang 1 , Ngu

Mike Flood Application for Solving Inundation

Issues for Ho Chi Minh City in The Context of

Climate Change: A Case Study in the District 1

Pham Thanh Long (1)(*) , Tran Tuan Hoang (1) , Nguyen Phuong Dong (1)

1 Sub-Institute of Hydrometeorology and Climate change, Ho Chi Minh City, Vietnam

* Correspondence: longpham.sihymete@gmail.com

Abstract: In recent years, urban flooding has caused significant impacts on people's activities,

socio-economic development, environmental pollution and traffic congestion in the Ho Chi Minh City There are more and more extreme rain events in particular climate change context, along with the impacts of sea level rise on the tides that increase the water level of canals and rivers, flooding has been occurred heavily, even the roads without any rains By using Mike flood model, intensity-duration-frequency (IDF) curve is used for calculating the flooding level and time, taking the terrain, drainage system, sewer and the size of a culvert in accounting for Ho Chi Minh City, a pilot for district

1 The results can be used for technical infrastructure planning, construction of flood warning and forecasting models, especially for flooding management in the connection with the smart city for Ho Chi Minh

Keyword: Inundation; Climate change; Mike flood; Ho Chi Minh City

1 Introduction

Over the years, the economic growth rate of Ho Chi Minh City (HCMC) has increased sharply, but not synchronously Many problems occur such as urban flooding, saline intrusion, environmental pollution, etc are mentioned in annual reports of departments The whole city has an area of 2061.4 km2 in 2016 but nearly 60% of the total area is located below 1.5 m above sea level So, the city is facing frequent flooding problems during the rainy season (from June to November each year) In the last decade, the urbanization rate of Ho Chi Minh City happened very fast with large works constantly appearing in the downtown area, especially in District 1 It leads to putting more pressure

on infrastructure (Larsen et al., 2008) as well as a drainage system (Nie et al., 2009, Mai et al., 2019) Consequently, the central roads have often been flooded when heavy rains occurred continuously

Currently, Vietnam is also one of the countries vulnerable to the impacts of climate change with the manifestation of extreme weather events increasing in frequency and unpredictable The frequency of appearance of heavy rains causes inundating the area (Tran

et al., 2015) more and more and directly affect traffic and people’s lives Urbanization of an area will, however, increase the magnitude of impacts on runoff and hydraulic performance (Tait et al 2008) There have many studies on related issues Grum et al (2006) simulated and analyzed the effects of rainfall on urban drainage systems using the ECM model Considering the impact of climate change on rainfall in an urban basin (Willems et al., 2012)) and establishing standards for design and operating urban drainage systems (Arnbjerg-Nielsen et al., 2013) is investigated Olofsson (2009) also analyzed the characteristics of the urban drainage system when considering climate change scenarios B2 and A2 Basing on

this base, the previous studies assessed the impact of climate change (Berggren et al., 2011; 2014) and provide a feasible approach for adapting with climate change to urban drainage systems (Neumann et al., 2015) The influence of urban development and global climate change on increasing rainfall intensity in HCMC area was also studied and discusses in an international Conference on the impact of climate change on Inundation (Luong 2008) Approaching of constructing the IDF curve based on extreme rain events according to each rain period was developed in the study of Nguyen et al (2017)

The paper aims to review, evaluate and determine the intensity of rain following the frequency of rain, serving the calculation of designed rain and floods for water drainage and urban planning works Therefore, we chose an approach that is to build Intensity - Duration

- Frequency (IDF) of rainfall at present and future climatic conditions under the impact of climate change Subsequently, developing future IDF scenarios to assess the degree of change in rain events and the impact of existing drainage systems in District 1, HCMC

2 Methodology

In this study, we apply the IDF curve to assess the intensity and frequency of rain for serving water drainage system in HCMC, the pilot area is District 1 Collected data include water level, meteorology-hydrology data, drain size data, etc and comprehensive inheritance of the database in the research area The models are used for simulating taking account for different scenarios with flooded periods such as MIKE FLOOD, MIKE 11 HD, MIKE 21 FM and MIKE URBAN (DHI, 2014) The ArcGIS 10.1 software suite is used to present District 1 flood calculation results on the base map

2.1 Establishing models and data

MIKE 11 HD model

- Boundary conditions (DHI, 2014):

o Upstream boundary: hourly water level boundary at stations: Tan An (Vam Co Tay river), Go Dau (Vam Co Dong river) from 0:00 on 01/01/2016 to 23:00 on 30/11/2016 and daily flow margin from 01/01/2016 to 30/11/2016 of the lakes: Dau Tieng (Saigon River) and Tri An (Dong Nai River)

o Downstream boundary: data of water level to be monitored at Vung Tau station from 0:00 on 01/01/2016 to 23:00 on 30/11/2016

MIKE URBAN model

From the collected data, filtered and removed some data such as small sewer lines, insufficient data Besides, the data needs to be homogenized with the coordinate system into WGS 84 UTM Zone 48 N (EPSG: 32648) After aggregating the complete data set, the digitization into MIKE URBAN is done by importing from the GIS data set via the Import tool (DHI, 2014)

To ensure the system works well with the model, it is necessary to check digitized data The cases leading to the model generating errors may include missing ground elevation data, lack of diameter data, the bottom elevation is higher than surface elevation,

the slope in the drain is unreasonable, the system is broken, etc “Project Check Tool” is used

as a tool to find out where the errors are Easy to fix errors and perfect the system after digitizing, including 1902 tunnel, 37 outlets, 1343 Round drain and 717 Square culverts

MIKE 21 FM model

Digital elevation data (LiDAR) is generated using the latest LiDAR technology of the Department of Science and Technology HCMC with a resolution of 5x5 m We need to convert LiDAR digital elevation data to * xyz file format in MIKE ZERO into the Mesh Generator tool which will be terrain in the model MIKE 21 FM

MIKE FLOOD model

Using MIKE FLOOD model connecting with MIKE 11 HD model, MIKE 21 FM and MIKE URBAN to calculate floods for District 1 area The paper made a connection at 1902 underground station and 37 outlets of the current status of District 1 drainage into the model follow these steps:

 Network connection in MIKE 11 HD model and terrain in the MIKE 21 FM model with Link river branch to Mike 21

 Connect river network in MIKE 11 HD model to 37 outlets of MIKE URBAN with Link river branch to MIKE URBAN tool

 Connect 1902 manhole from MIKE URBAN model to MIKE 21 FM model terrain using Link urban node to MIKE 21

Figure 1 Diagram of the river network in

hydraulic model MIKE 11 HD

Figure 2 Current situation of drainage in District 1 in the MIKE URBAN model

Figure 3 The topography of District 1 in

MIKE 21 FM model

Figure 4 Diagram of the model connection between modules (MIKE 11 HD, MIKE 21 FM and MIKE URBAN) in MIKE FLOOD model

interface

2.2 Design rain chart

The design rain chart is built for a period of 180 minutes, impulse rain 5 minutes /1 data The rain chart was built by the method of instant rain intensity based on the delegate rain (September 26, 2016, the amount of rain was 132 mm, 3-hour period) causing heavy flooding in District 1 Combining with IDF curve of rain is built for Ho Chi Minh City according to the short-term rain data at Tan Son Hoa station (1971-2016) and the IDF curve

is constructed according to a future scenario under climate change condition Design rain charts are used as the input rain margin for the MIKE URBAN model

Figure 5 IDF line for rainfall based on short-term

monitoring data (1971-2016) at Tan Son Hoa station

Figure 8 Representative rainfall chart of September 26, 2016 (3-hour period)

Figure 6 IDF road rains under average scenario at

Tan Son Hoa station

Figure 9 Design rain chart of the 10-year repeated rain cycle according to the average scenario (3-hour period)

Figure 7 IDF road under high scenario rain at Tan

Son Hoa station

Figure 10 Design rain chart of the 10-year repeated rain cycle according to the high scenario (3-hour period)

2.3 Develop flood calculation scenarios

During the rainy season, to ensure proper inter-reservoir operating procedure and safe for the lake, regulatory authorities will release regulated water through the spillway Under the impact of climate change, this overflow will change So, in this study, the upstream boundaries are assumed to be the running volume plus the overflow discharge at Dau Tieng reservoir and Tri An has an increase in rainfall flow

The selected scenarios to calculate for the 2016 status quo and the proposed scenarios include: Medium and high scenario for 2030 (early century) corresponding to the 10-year repeated rain period (According to Scenario of Climate Change and Sea Level Rise for Vietnam developed by the Ministry of Natural Resources and Environment with the latest version in 2016) The applied flood calculation for District 1 area is presented in Table 1

Table 1 Model calculation scenarios

No Scenarios Period Water level

downstream

Precipitation on-site in Ho Chi Minh City area

Upstream flow

2016

Rain chart designed for the delegate rain on September

26, 2016 (rainfall 132 mm)

Data of Ho Dau Tieng Reservoir

Tri An in 2016

RCP4.5

2030 early of the century

Increase 12 cm Rain chart designed

according to the early stages

of the century + Repeated cycle 10 years (160.5 mm rainfall)

Dau Tieng increase: 33

m3/s

Ho Tri An increase: 90

m3/s

RCP8.5

2030 early of the century

Increase 12 cm Rain chart designed

according to the early stages

of the century + Repeated cycle 10 years (precipitation 178,5 mm)

Dau Tieng increase: 30

m3/s Tri An increased: 80

m3/s

3 RESULTS

3.1 Calibration results and test the models

MIKE 11 HD model

Based on the input data set and input conditions, boundary conditions have been established, the set of parameters of the 1-D flow model is calibrated with the actual water level data measured at Phu An station, Nha Be from 9:00 September 28, 2015 to 09:00 hours September 30, 2015 With the Nash index all over 0.90

Figure 11 The water level of Phu An station

September post-calibration (Nash = 0.97)

Figure 12 The water level of Nha Be station in September after calibration (Nash = 0.98)

To test a hydraulic model, we use a modified set of parameters and change the calculation time Testing time: 3 days from 9:00 September 17, 2016 to 9:00 September 19,

2016 Model test results are also evaluated by Nash index above 0.90

Figure 13 The water level of Phu An station

in September after a test (Nash = 0.98)

Figure 14 The water level of Nha Be station in September after a test (Nash = 0.98)

MIKE URBAN model

In the MIKE URBAN model, the parameters used to adjust consisting of a width of the manhole, sewer diameter, as well as the size of the drain (length and area of the section), Manning’s roughness coefficient, loss coefficient, impermeability coefficient of sub-basin collecting water Through the rain on September 26, 2016, the model was adjusted to the above parameters to get results consistent with the reality For residential areas, the waterproof coefficient is 65% and roads are 80%

3.2 Results of submergence using MIKE FLOOD model

Corrected flood calculation results

According to 2016 statistics in District 1 flooded areas caused by extreme rain, mainly extreme rain on September 26, 2016 (132 mm arms, 3-hour period exceeding the design standards of HCMC drainage system according to Decision 752 / QD-TTg with 95.91

mm, 3-hour period) The depth of flood results calculated from the model is compared with the statistics of flood spots (from the Executive Center of the Ho Chi Minh City Flood Control Program) shows The model results are quite consistent with the current rain on September 26, 2016 (presented in Table 2)

Table 2 Comparison of the flooded depth for model and observation in District 1

Numerical

order

Street names

Flooded location Actual

flood depth (m)

Calculated flooded depth (m)

Quynh

Thai Binh Market Nguyen

Cu Trinh

Cu Trinh

Tran Hung Dao Tran Dinh

Xu

Hung Dao

Ho Chi Minh City Police Headquarters

Nguyen Van Cu

Luu

5 Nguyen

Van Cu

Results of flood calculation in District 1 according to the scenarios

According to 2016 flood

statistics, District 1 has about 5

flooding points Floodplains are

concentrated on Nguyen Thi Minh

Khai roads, Mai Thi Luu, Nguyen Cu

Trinh, Tran Hung Dao and Nguyen

Van Cu Compared with the results of

flooding, the current situation is

relatively accurate with the depth of

flooding fluctuating 0.2 ~ 0.4 m

Flooding time is 60-90 minutes

The paper continues to perform

flood simulation in the study area of

District 1 under future scenarios for

2030 with a 10-year repetition cycle

under two medium scenarios (design

rainfall 160.5 mm) and high scenario

(design rainfall is 178.5 mm) The

results of flood simulation according to

the scenarios are detailed as shown in

Figure 16, Figure 17:

Figure 15 Flood map of District 1 area in

2016

According to the calculation results in 2030, in both RCP4.5 and RCP8.5 scenarios, the level

of flooding increased significantly compared to the simulation results of the 2016 rainfall The depth of flood ranges from 0.2 ~ 0.53 m, flooding time is 70 - 100 minutes (according to the medium scenario) The depth of immersion ranges from 0.25 ~ 0.6 m, flooding time is 70

- 120 minutes (according to the high scenario)

Figure 16 The highest flood map for District 1

in 2030 under the medium scenario

Figure 17 The highest flood map for District 1

in 2030 under the high scenario

Through analysis of flood results in RCP4.5 and RCP8.5 scenarios at the beginning

of the century, corresponding to the design rain according to the 10-year repetition cycles, shows:

 The situation of flooding of District 1 roads increased Specifically, the flooded area increases with each scenario The area flooded with rain on September 26, 2016 was 12.83 hectares, according to the 10-year cycle of repeated rains in the average scenario is 17.74 ha and according to the 10-year cycle of repeated rain in the high scenario, it is 20.77 ha (see Table 3)

Table 3 Results of District 1 flooded area statistics according to the calculated scenarios

Repetitive

scenario/period

of rain

Flooded area according to flood level (ha) Area

(ha)

Flood rate (%) 0,1-0,15

(m)

0,15-0,2 (m)

0,2-0,25 (m)

0,25-0,3 (m)

0,3-0,5 (m)

0,5-0,8 (m) Status quo 3,68 2,99 2,58 1,41 2,17 0,22 12,83 1,66 Medium

scenario

High scenario 6,06 4,48 3,86 2,61 3,75 0,49 20,77 2,69

 Extreme rain in the future will reduce the responsiveness of many tertiary sewers, some secondary sewers, as a result of the statistics, the flooded area of the County leads to the effect of drainage

 Extreme rain has increased, long flood time, high flooding has affected traffic and drainage capacity of the whole District 1 drainage system

4 Proposed Solutions

In fact, due to rain in urban areas, the buffer element plays a very important role The park or vacant land is the ideal place to drain rainwater from which will reduce the possibility of flooding due to rain in the city in general and of District 1 in particular The rest are residential areas, or road surface, pavement… mainly made of concrete, brick, rock

so poor water permeability, encountering heavy rainfall and short time (extreme rain) would be very easy to flood In the future, District 1 will invest in developing waterproof bricks, for use as a walking pavement, this contributes to reduce flooding and increase drainage capacity for District 1

After adjusting the permeability coefficient suitable for works such as pavement, which has been linked with bricks, it is good waterproof, the heavy flooding points have reduced as in the Nguyen Thi Minh Khai section, the flooded area has decreased from 211

m2 to 89 m2 The sections of Mai Thi Luu and Cong Quynh were no longer flooded, the average depth of inundation decreases from 0.5 m to 0.3 m

Figure 18 Flooding after the rain on

September 26, 2016

Figure 19 Flood District 1 after using absorbent brick lining solution for sidewalks

5 CONCLUSIONS

Results from the model for rain event on September 26, 2016 with the highest depth

of 0.41 m, the inundated area is 12.83 ha For future design rains with a 10-year repeat period

of medium and high emission scenarios, at the beginning of the century, the situation of flooding in District 1 is more complicated Inundated time increases, the highest depth of