[PPT] - Wet weather water quality monitoring and urban flood analysis in PowerPoint Presentation

SLIDE 1

Wet weather water quality monitoring and urban flood analysis in Hue Citadel area

Hiroaki Furumai

Professor Research Center for Water Environment Technology Department of Urban Engineering The University of Tokyo

SLIDE 2

2

Canal-pond network and Huong river

The Hue Citadel area is located at

12km upstream from the Huong river mouth. Area: 5.21km2 Population: 63,638

The inner canal is linked with the
uter canal which is connected

with Huong river.

In rainy season, inundation occurs

several times a year. 1km

N

Citadel area

GEOSS/AWCI: May 27, 2014

Inundation situation

n Nov. 16th 2013

Le Thanh Ton Street

SLIDE 3

1. To investigate the characteristics of water

pollution focusing on fecal contamination in the canals and ponds during dry and wet weather periods

2. To assess the influence of river water inflow and

wastewater discharge on water flow and water pollution in canals by continuous monitoring with water depth and EC sensor

3. To develop urban inundation model considering

river water level change

3 GEOSS/AWCI: May 27, 2014

Research Objectives

SLIDE 4

GEOSS/AWCI: May 27, 2014 4

Water sampling and Continuous monitoring

*JICA(2006), **Lieu et al(Hue University)

B1 B2 B3 P1 P2 P3 P4 P5 P7 C1 C2 C3 S1 S2 S3 S4 1km

N Water sampling (16 points) was conducted during dry and wet weather in 2012. Water quality parameters : are E.coli, Total coliform, COD, NH4- N, EC etc. :Canal (B,C) :Pond (P) :Street（S） Continuous monitoring (13 points)

was conducted by water depth and EC sensors during a rainy season, Sep. to Dec. in 2012.

Water depth sensor EC sensor Vinyl pipe * Measured water depth

SLIDE 5

GEOSS/AWCI: May 27, 2014 5

E.coli and Total coliform (TC) concentrations of most samples exceeded the

standard values (23 and 24 out of 27 samples (85 and 89%), respectively)

E.coli and TC concentrations were higher in many samples than the regulation

values (5 and 8 out of 8 samples, respectively).

Inundated water samples at streets also showed high E.coli concentration.

Fecal contamination in canals, ponds, and inundated water

10 100 1000 10000 100000 Canal Pond Street 104 103 102

E.coli (CFU/100mL)

N.D. 105

QCVN:B2

{9, 10, 0} {4, 4, 4} n＝

Dry weather Wet weather

100 1000 10000 100000 1000000 10000000 Canal Pond Street

Total Coliform (CFU/100mL)

N.D. 106 105 104 103 107

QCVN:B2

{9, 10, 0} {4, 4, 4} n＝

P7 (Sep.) C2 (Sep.) (Dry weather sample) (Wet weather sample)

DL 102

SLIDE 6

GEOSS/AWCI: May 27, 2014 6

Use of EC as pollution indicator and EC monitoring in canal

96 142 187 264 450 427 396 352 163 158 69 72 129 241 234 190 152 135 58 135 56 100 200 300 400 EC (μS/cm)

1km

N

y = 0.529x + 9.2894 R² = 0.9976 50 100 150 200 250 300 200 400 600

EC (μS/cm) TDS (mg/L)

EC vs TDS

y = 0.0026x + 0.04 R² = 0.8096 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 200 400 600

EC (μS/cm) NH4-N (mg/L)

EC vs NH4-N EC vs Total coliform

y = 0.002x + 3.997 R² = 0.4735 3 3.5 4 4.5 5 5.5 200 400 600

EC (μS/cm) Total coliform (log(CFU/100mL))

QCVN:B2

SLIDE 7

Inundation simulation in Hue City

Model simulation has been conducted in the Hue Citadel area to explain the inundation situation. We plan to conduct model simulation under climate change and discuss on possible effective countermeasures for river flood and urban inundation control.

Simulation results Past inundation record

GEOSS/AWCI: May 27, 2014 7

SLIDE 8

GEOSS/AWCI: May 27, 2014 8

・Sewer system

1) Drainage system data

・River, channel and pond ・Ground elevation

2) Ground elevation and surface data

・Land use ・Rainfall

3) Meteorological data

・Water level of river, channel or pond

4) Hydrological data

・Inundation depth and area ・Water quality of inundated water

5) Data for model calibration

Required data for model simulation

SLIDE 9

GEOSS/AWCI: May 27, 2014 9

Data collection Model construction Inundation simulation Model calibration

For model construction

Drainage system
Land use
Ground elevation

Data collection

For parameters of calculation

Rainfall data
Water level and quality of river
Water level and quality of

channels and ponds in urban area

Flow and quality of wastewater

Scenario analysis

Data collection

For calibration

Record of inundation area, depth
Water quality of inundated water

Data processing by GIS software Checking the model performance

Flowchart of urban inundation model development and Scenario analysis

Predicted future rainfall data and river flow data under climate change Possible measures for flood and pollution control

SLIDE 10

Inundation simulation with water level rise

10

0.5 0.1 Inundation depth(m) 0.2

Water level 2.0m ： inundation area = 232ha Water level 1.0m : inundation area = 172ha Freefall ：inundation area = 167ha

Inundation area expands with increasing the river water level. Therefore, it is very important to consider river water level to evaluate urban inundation situation. Simulation condition:

Return period 2 years -2 days rainfall Maximum rainfall intensity: 61.1mm/hr Total rainfall: 164mm

10 20 30 40 50 60 70 80 1 5 9 13 17 21 25 29 33 37 41 45

Duration (hours) Rainfall intensity(mm/hr) 2year-2day rainfall (interval 60 minutes)

Total: 164mm

61.1mm/hr

GEOSS/AWCI: May 27, 2014

SLIDE 11

Increase of heavy rainfall event in Hue

GCM model prediction on rainfall -

GEOSS/AWCI: May 27, 2014 11

2 4 6 8 10 2 4 6 8 10

> 50mm/d > 100mm/d > 200mm/d > 50mm/d > 100mm/d > 200mm/d

1981-2000 2046-2065

[days] [days] 2.40 7.66 0.76 2.28 0.54 6.83

5 GCM models : GFDL, MIROC_H, MIROC_M, MIUB, GISS
Model calculation (1981-2000) and prediction (2046-2065)
Average frequency of 5 models during rainy season (Sep-Dec)

SLIDE 12

GEOSS/AWCI: May 27, 2014 12

Fecal pollution: Most of the canals and ponds are fecally

contaminated during both dry and wet weather periods.

EC usage as an indicator of water flow: EC value indirectly

indicates pathogenic pollution level. EC sensor is very effective tool to know continuous change of water flow as well as water pollution in urban canal which is affected by wastewater and river water inflow.

Urban inundation simulation: The developed sewerage

model linked with river water level is useful to estimate the detailed inundation characteristics in the Citadel area. In the next step, it is necessary to conduct model estimation

f pathogenic pollution during rainy season and to

propose effective measures for flood and health risk management under climate change.

Conclusions and future task

SLIDE 13

13

GEOSS/AWCI: May 27, 2014

Thank you for your attention

Hiroaki FURUMAI

Professor, Research Center for Water Environment Technology, Graduate School of Engineering, University of Tokyo furumai@env.t.u-tokyo.ac.jp

River flood Inland flood (Inundation)

SLIDE 14

Key Points of Inter-linked Research in Hue

3 Risks: flood risk, inundation risk, health risk

Hydrological model --- Watershed scale Inundation & runoff quality model --- Drainage scale Health risk model --- Community/Human scale

3 M: Modeling, Monitoring, and Management

Data collection for Modeling Model calibration and validation Sampling and Monitoring work Scenario development and analysis for Management Multi-scale analysis Inter-linked research

GEOSS/AWCI: May 27, 2014 14

Adaptation to Climate Change

=> Understanding and assessing climate change-related risk

SLIDE 15

Inter-linked Cooperative Research in Huong River Basin

Research Steps and Final Goal

GEOSS/AWCI: May 27, 2014 15

How to manage river flood, urban inundation, and flood-related health risk under climate change?

WEB-DHM model --- Watershed scale (River flood risk) WQ Monitoring ---- Drainage scale (Water quality risk) Urban inundation model --- Drainage scale (Inundation risk) Dose-response relationship --- Community/human scale (Health risk)

1) Evaluation of river flow/floods at the Hue city using the watershed hydrological model. 2) Evaluation of inundation characteristics using the urban inundation model considering the output of 1) . 3) Evaluation of pathogenic pollution during flooding based on the output of 2) and water quality monitoring. 4) Estimation of human health risk during flooding.

SLIDE 16

GEOSS/AWCI: May 27, 2014 16

SLIDE 17

River flood

GEOSS/AWCI: May 27, 2014 17

River flooding in Hue city at Nov. 8th, 2013. http://talkvietnam.com/2013/11/hue-city- hydropower-dams-open-districts-submerged/) Inland flood (inundation) in Hue city at Sep. 4th, 2009. http://tropical.way-nifty.com/blog/2009/09/rainy- season-ha.html)

Inland flood

SLIDE 18

18 GEOSS/AWCI: May 27, 2014

Research Background

flood and water-born diseases -
Frequent urban flooding and its

damage during rainy season in Southeast Asia

Flooded period

Epidemic period 90 % percentile

Mean number

f cases per week

Weeks Flood starts

Schwartz et al., 2006

High occurrence rate of water-

born diseases during and after urban flood

Inland flood (inundation) in Hue

n Sep. 4th, 2009.

SLIDE 19

GEOSS/AWCI: May 27, 2014 19

Rainy season is Sep. to Dec.
Serious inundations occurred in

1999 and 2004. Inundation depth exceeded 2m.

978mm 682mm

200 400 600 800 1000 1月 2月 3月 4月 5月 6月 7月 8月 9月 10月 11月 12月

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005

Rainfall(mm/year)

1999 2004

Jan. Feb. Mar. Apr. May. Jun. Jul. Aug. Sep. Oct. Nov. Dec.

Inundation depth in 1999 Nov.

Inundation Depth

Inundation situation

Monthly rainfall from 1995-2005

Rainy season Dry season

SLIDE 20

Limited capacity of urban drainage with pollution

In heavy rainfall events, wastewater and runoff water have chance to overflow to streets in the downtowns, because the drainage ca

pacity was designed for the less-developed condition. In

addition, lakes and rivers have been polluted both by wastewater discharge from sewerage system and non- point pollution sources such as wash-off from surface land.

Importance of pathogenic pollution monitoring

It is meaningful to investigate wet weather pollution in urban runoff, overflow to lakes from sewer system and inundated water, focusing on pathogenic indicators.

20 GEOSS/AWCI: May 27, 2014

Research Background (3)

Limited drainage capacity and monitoring data -

SLIDE 21

C1 C2 C3

1km

N

GEOSS/AWCI: May 27, 2014 21

EC change during dry weather

C1 and C3 are affected by the river water inflow from river water. At C2, canal water might be stagnant and less diluted by river water.

Dry weather period

Introduction Methodology Results Summary

C1 C3

100 200 300 400 500 0.2 0.4 0.6 0.8 1 1.2 9/17 9/17 9/18

Water depth* (m) EC (μS/cm)

0:00 12:00 0:00 100 200 300 400 500 0.2 0.4 0.6 0.8 1 1.2 9/17 9/17 9/18

Water depth* (m) EC (μS/cm)

0:00 12:00 0:00

C2

100 200 300 400 500 0.2 0.4 0.6 0.8 1 1.2 9/17 9/17 9/18

Water depth* (m) EC (μS/cm)

0:00 12:00 0:00

Water depth

EC

EC ranges; River water: 40 to 70 μS/cm (*2) Wastewater: 660 μS/cm (*1)

(*1): Feb. 2012 sampling (*2): Report from HueWACO (water supply company)

SLIDE 22

C1 C2 C3

1km

N

GEOSS/AWCI: May 27, 2014 22

EC change during wet weather

Wet weather period

At C1 and C2, canal water was diluted by the river water inflow. C3 might be affected by both the polluted inner canal water and the cleaner river water.

Introduction Methodology Results Summary

100 200 300 400 500 0.2 0.4 0.6 0.8 1 1.2 10/6 10/7 10/7

Water depth* (m) EC (μS/cm)

Water depth EC

12:00 0:00 12:00

C1 C2 C3

100 200 300 400 500 0.2 0.4 0.6 0.8 1 1.2 10/6 10/7 10/7

Water depth* (m) EC (μS/cm)

12:00 0:00 12:00 100 200 300 400 500 0.2 0.4 0.6 0.8 1 1.2 10/6 10/7 10/7

Water depth* (m) EC (μS/cm)

12:00 0:00 12:00

SLIDE 23

Research objectives

To develop a urban inundation model by collecting

data of drainage system, land use and ground elevation etc.

To simulate the inundation situation by the urban

model considering Huong river water levels

To discus on pathogenic pollution during

flooding/inundation period

To estimate the future inundation characteristics

under climate change

To propose effective measures by scenario analysis

23 GEOSS/AWCI: May 27, 2014

SLIDE 24

GEOSS/AWCI: May 27, 2014 24

1D water flow analysis during dry weather Appendix 2.1

Water level Time B3 B1

C3 C2 C1

1D analysis Condition: Water levels are same in high tide

Low tide High tide High tide Low tide

SLIDE 25

C1 C2

C3

1km

N

C1 C2

C3

1km

N

50 100 150 200 250 300 350 400 450 500

10
5

5 10 15 20 10/6 10/7 10/7

EC Water flow

12:00 0:00 12:00

Water flow (m3/s) EC (μS/cm)

GEOSS/AWCI: May 27, 2014 25

1D water flow analysis during wet weather Appendix 2.2

1D analysis was applied to wet weather period. Flow direction might explain why EC fluctuated at C3.

SLIDE 26

C1 C2 C3

1km

N

GEOSS/AWCI: May 27, 2014 26

Wastewater stagnation during dry weather

C1 and C3 are affected by the river water inflow from river water.
At C2, canal water might be stagnant and less diluted by river water.

Introduction Methodology Results Summary

100 200 300 400 500 600

0.15
0.1
0.05

0.05 0.1 0.15 17-Sep 17-Sep 17-Sep

Water level change (m) EC (μS/cm)

6:30 12:30 18:30

Water level change at C1

EC C2

C1 C3

EC ranges; River water: 40 to 70 μS/cm (*2) Wastewater: 660 μS/cm (*1)

(*1): Feb. 2012 sampling (*2): Report from HueWACO (water supply company)

Continuous monitoring was conducted from Sep. 2012 in the canals.

SLIDE 27

C1 C2 C3

1km

N

GEOSS/AWCI: May 27, 2014 27

Dynamic EC change during wet weather

At C1 and C2, canal water was diluted by the river water.
C3 might be affected by both the inner canal water and the river

water.

Introduction Methodology Results Summary

100 200 300 400 500 0.1 0.2 0.3 0.4 0.5 0.6 0.7 6-Oct 7-Oct 7-Oct

Water level change (m) EC (μS/cm)

12:00 0:00 12:00

Water level change at C1

EC

C2 C1

C3

SLIDE 28

Required data for inundation simulation

GEOSS/AWCI: May 27, 2014 28

1) Drainage system data

Drainage system data was made by JICA in 2006. Size and location of ponds were given by Lieu. (Hue Univ.)

Conduits Ponds Manholes Canals

Ground

Ground Elevation

Spill crest Invert elevation Bottom elevation Diameter Length Shape

Manhole Conduit

SLIDE 29

GEOSS/AWCI: May 27, 2014 29

2) Ground elevation and surface data

Hue Citadel area

Ground slope data (TIN; Triangular Irregular Network) was made from ground elevations of 576 nodes. (495/576 were manholes)

Additional canals and ponds Radar elevation survey device on a car (TOPCON)

Elevation survey using radar was done on August 2012.

Required data for inundation simulation

SLIDE 30

GEOSS/AWCI: May 27, 2014 30

Required data for inundation simulation

2) Ground elevation and surface data

Hue Citadel area

Road land use data was made by JICA.

Vegetation layer

(NDVI + NVEI）

IKONO satellite image

Multi spectrum image- red/blue/green/NIR

Land use can be evaluated by satellite image data.

SLIDE 31

GEOSS/AWCI: May 27, 2014 31

・Rainfall Required data for inundation simulation

3) Meteorological data

1 hour interval data measured by Hue Meteorology and Hydrology Center.

20 40 60 9/1 10/1 10/31 11/30 12/30 1/29 Rainfall (mm/hr)

Year of 2011

・Water level of river at Kim Long station

4) Hydrological data

1 hour interval data measured by Hue Meteorology and Hydrology Center.

SLIDE 32

GEOSS/AWCI: May 27, 2014 32

At 16 points, including 3 points on street, water level and EC sensors are installed.

165mm 32mm

EC logger

150mm 25mm

Water level logger

B1 B2 B3 P1 P2 P3 P4 P5 S1 S2 P7 S3 P6 C1 C2 C3

Measured parameters by sensors Water level Water level・EC (and Turbidity at C3) P:Pond C:Canal B:Boundary between Canal and Huong river S:Street（For Inundated water）

Required data for inundation simulation

Hue Citadel area

Survey for setting the sensors （At S2, frequently inundated point）

5) Data for model calibration

・Water level and water quality of inundated water

Sensors are useful to get WQ data under different inundation conditions.

SLIDE 33

GEOSS/AWCI: May 27, 2014 33 1km

0.5

Depth (m)

0.2

No consideration of river water level Flooded area: 82.6ha Max depth: 0.95m Flooded area: 100.2ha Max depth: 0.96m Given with river water level: 2.0m

1D2D analysis by XPSWMM

Simulation condition:

Return period 2 years -2 days rainfall Maximum rainfall intensity: 61.1mm/hr Total rainfall: 164mm

Urban inundation simulation

10 20 30 40 50 60 70 80 1 5 9 13 17 21 25 29 33 37 41 45

Duration (hours) Rainfall intensity(mm/hr) 2year-2day rainfall (interval 60 minutes)

Total: 164mm

61.1mm/hr

SLIDE 34

Collaboration with other GRENE groups

in the simulation of future flooding situations, need

prediction data of rainfall and fluctuation of water level.

– If rainfall predictions are different in the upstream and downstream of Huong river, must consider flow rate Q (m3/sec)

f whole Huong river basen.
Corroborate with calculate result of other GRENE groups!

34

Thua Thien Hue Province (Phong Tran et al. 2007.) Hue city → (H-Q curve of Huong River observed in Kim Long Station, 2005.) ↑ Kim Long Station

H = 1.2438xQ0.6908 R2 = 0.8267 50 100 150 200 250 300 350 1000 2000 3000 4000 H (Cm) Q (m3/s)

Relationship Q=F(H) - Kim Long

GEOSS/AWCI: May 27, 2014

SLIDE 35

List of required data to simulate with the distributed model

1. Drainage system data

– Sewer system – River, canal and pond

2. Ground surface data

– Ground elevation – Landuse

3. Meteorological data

– Rainfall

4. Hydrological data

– Water level of river, canal and pond

5. Calibration data

– Water level of canal or pond – Inundation depth and area

GEOSS/AWCI: May 27, 2014 35

2. Methodology
Num. of Nodes : 495
Num. of Links : 677

from JICA in 2006.

Ground

Ground Elevation

Spill crest Invert elevation Bottom elevation Diameter Length Shape

node links

SLIDE 36

List of required data to simulate with the distributed model

1. Drainage system data

– Sewer system – River, canal and pond

2. Ground surface data

– Ground elevation – Landuse

3. Meteorological data

– Rainfall

4. Hydrological data

– Water level of river, canal and pond

5. Calibration data

– Water level of canal or pond – Inundation depth and area

GEOSS/AWCI: May 27, 2014 36

2. Methodology

SLIDE 37

List of required data to simulate with the distributed model

1. Drainage system data

– Sewer system – River, canal and pond

2. Ground surface data

– Ground elevation – Landuse

3. Meteorological data

– Rainfall

4. Hydrological data

– Water level of river, canal and pond

5. Calibration data

– Water level of canal or pond – Inundation depth and area

GEOSS/AWCI: May 27, 2014 37

2. Methodology

Rainfall (mm/hour) Water level (m)

SLIDE 38

List of required data to simulate with the distributed model

1. Drainage system data

– Sewer system – River, canal and pond

2. Ground surface data

– Ground elevation – Landuse

3. Meteorological data

– Rainfall

4. Hydrological data

– Water level of river, canal and pond

5. Calibration data

– Water level of canal or pond – Inundation depth and area

GEOSS/AWCI: May 27, 2014 38

2. Methodology

Past inundation record Simulated area

SLIDE 39

GEOSS/AWCI: May 27, 2014 39