The after-runner storm surge along the north coast of Vietnam simulated by the 2D ROMS model

Nguyen Ba Thuy1, Pham Khanh Ngoc1, Lars R. Hole2, Nils Melsom Kristensen2, Johannes Röhrs2, Cecilie Wettre2

1Vietnam National Center for Hydro-meteorological Forecasting (NCHMF) 2Norwegian Meteorological Institute (MET Norway)

Cape Town, 3-7 April 2017

Storm surge

Anormalyze water level (Meteorological tsunami)

Spring tide (Hight tide

+ surge induced by moonson)

Coastal Dissater related to high water level in Vietnam

High Vulnerable storm surge (abundant typhoon +

shallow water)

Less Vulnerable storm surge (rare typhoon + deep water)

Less Vulnerable storm surge

(shallow water +rare typhoon)

Number of

Typhoon

Possibility Storm Surge in Vietnam

June

Dec.

-0.5

0.0

0.5

1.0

1.5

9/29/2006 6:00 9/30/2006 12:00 10/1/2006 18:00

Sur

ge h

eigh

t (m

)

Time (h)

Observation

Without wave

With waveNumerical Model:

- 2D ROMS + WRF

- SuWAT (Surge Wave and

Tide) + Best track or WRF

Time serial of storm surge

Storm surge research and prediction

Coastal inundation due to tide + storm surge + wave setup

Flooding in Ho Chi Minh City in

20/11/2011 Maximum tide amplitude

Spring tide in Hochiminh city

Spring tide in Hochiminh City is related to Surge induced by

monsoon or not?

It related to flooding in Hochiminh

City on 20/10/2013

-0.5

-0.3

-0.1

0.1

0.3

0.5

0.7

0.9

10/8/2013 0:00 10/14/2013

0:00

10/20/2013

0:00

10/26/2013

0:00

Time (s)

Su

rge

hei

ght

(m

)

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

Wat

er s

urf

ace

elev

atio

n (

m)

OBS-Total water levelModel-SurgeOBS-Surge

Spring tide in Hochiminh city on 20/10/2013

* Vung Tau

Hochiminh city

Time profile of total water surface elevation, predicted and measure

surge height

Maximum water surface elevation during the monsoon 10/2013

Wind field at 20/10/2013 Spring tide = high tide+ surge (by moonson)

High Oscillation of water level on the middle coast of Vietnam

High tide at Tuyhoa in Dec. 2016

(No typhoon and calm wind. But the

inundunation height up to 1m, and go in

land up to 200m). Meteorological

tsunami?.

Time serial of total water level, tide and surge

To simulate this phenomenal need high resolution of ocean and atmosphere model. Next study

-1

0

1

2

3

4

5

6

12/11/2016 0:00 12/16/2016 0:00 12/21/2016 0:00

Z (c

m)

Time (h)

OBS Tide Surge

The after runner storm surge on the North coast of Vietnam

Area: The north coast of Vietnam

Typhoon frequency: 1.5 typhoons/year

Maximum storm surge: 3.4m

The after runner storm surge: 6 cases/50 years

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

1.2

9/16/2014 0:00 9/18/2014 0:00

Stoe

m su

rge

(m)

Time (h)

Storm surge at Hondau station during typhoon Wukong (2005) and Kalmaegy (2016)

Typhoon landfall

Typhoon landfall

Normal storm surge

After runner storm surge

The after runner storm surge

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

1.2

7/29/2005 0:00 7/31/2005 0:00 8/2/2005 0:00

Stor

m s

urge

(m)

Time (h)

Observation water level, predicted tide and

storm surge at Hondau station

-50

0

50

100

150

200

250

300

350

400

450

9/16/2014

9:36

9/16/2014

14:24

9/16/2014

19:12

9/17/2014

0:00

9/17/2014

4:48

9/17/2014

9:36

9/17/2014

14:24

9/17/2014

19:12

9/18/2014

0:00

9/18/2014

4:48

9/18/2014

9:36

9/18/2014

14:24

Time (h)

H (

cm)

Total water level

Prediced tide

Surge

Time of typhoon landfall

time interval of

urge exceed 0.5m

The after runner of storm surge cases

Kalmeagi (9/2014)

Mirinae (7/2016)

Typhoon land fall

Inundated

No inundation

- Damage by wind is not much.

- But Storm surge + high tide + high wave

generated inundation at Haiphong coast

after the typhoon landfall 5 hours.

(a)

Typhoon Kalmaegi (9/2014) – After runner storm surge case

High wave and inundation at Haiphong coast

- Empirical Typhoon Model:

fDuy

Dgy

Dv

x

Duv

t

Dv

fDvx

Dgy

Duv

x

Du

t

Du

byay

ww

bxax

ww

)(1)(1

)(1)(1

02

02

0

y

Dv

x

Du

t

2

0)/(1

)(rr

PPPrP c

(Fujita model)

r

Pfv

r

v

12

Operational Forecasting Model in 9/2014

Two dimensional long wave model + Empirical Typhoon Model

Result of Operational forecasting (2D long wave model + Fujita model)

Simulated result showed storm surge not high at Hondau station, and at low tide –> No warning. But infact inundation occurred. We miss warning

Maximum storm surge

Model and observation storm surge

Typhoon track

-1

-0.5

0

0.5

1

1.5

2

9/16/2014 12:00 9/17/2014 12:00 9/18/2014 12:00

Stor

m s

urge

(m

)

Time (h)

OBS

FUJITA

Typhoon Landfall

Storm surge observation data

Time profile of total water, tide and storm surge at Hondau and Honngu station

-100

-50

0

50

100

150

200

250

300

350

9/16/2014 0:00 9/16/2014 18:00 9/17/2014 12:00

Z (

m)

Time (h)

OBS

Prec. Tide

Storm surge

Typhoon Landfall

-50

0

50

100

150

200

250

300

350

400

450

9/16/2014 12:00 9/17/2014 12:00 9/18/2014 12:00

Z(m

)

Time (h)

OBS

Predic. Tide

Storm surge

Typhoon land fall

Simulated the after runner storm surge by (ROMS 2D + WRF)

Domain for WRF model (7km) and ROMS model (2km)

WRF

ROMS

Results of 2D ROMS + WRF model

Comparison of calculated and simulated storm surge at two station using wind and pressure from WRF

ROMS+WRF is well simulated the storm surge at two station -0.6

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

9/16/2014 12:00 9/17/2014 6:00 9/18/2014 0:00

Stor

m su

rge

(m)

Time (h)

OBS

WRF

Typhoon landfall

-1

-0.5

0

0.5

1

1.5

2

9/16/2014 12:00 9/17/2014 12:00 9/18/2014 12:00

Sto

rm s

urge

(m

)

Time (h)

OBS

WRF

Typhoon Landfall

Measured and predicted pressure (above) and

wind speed (below) and at Hondau station

The WRF model simulated well wind and pressure than the

empirical typhoon model

Wind and pressure data

Comparison of calculated and simulated storm surge at two station for two case using WRF model. One using front and tail wind field (above), the other using the front wind field only (below).

Effect of front and tail wind field on storm surge

Storm surge at Hondau station is mostly induced by tail wind. But at Honngu station is due to the inertia force

-1

-0.5

0

0.5

1

1.5

9/16/2014 6:00 9/17/2014 6:00 9/18/2014 6:00Sto

rm s

urg

e (m

)

Time (h)OBS

WRF

WRF-No wind after land fall

No wind after typhoon land fall

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

9/16/2014 0:00 9/17/2014 0:00 9/18/2014 0:00Stor

m s

urge

(m

)

Time (h)

OBS WRF WRF-No wind after landfall

The result at the Honngu station is not so much different between two cases. It means storm surge at Honngu station is due to water moving from North to the south ?.

Maximum storm surge for the case the model using front and tail wind field (left) and front wind field only (right)

Effect of front and tail wind field on storm surge

Honngu

Hondau

At the time typhoon land fall (water level at two station is down) After typhoon land fall 4 hour (water level at Hondau station is going up)

After typhoon land fall 6 hour (water level at Hondau station get maximun)

The tail wind fields of typhoons are more important than front wind fields.

How the surge generated at the Northern part (Hondau station)

Wind and water surface elevation

After typhoon land fall 4 hour (water level at Hondau station is going up)

Water surface elevation and current

Storm surge on the southern part (Honngu station) due to the tail wind

and water moving from the north.

How the surge generate at the Southern part (Honngu station)

Wind and water surface elevation

After typhoon land fall 8 hour

Effect of wind and pressure on storm surge

Comparison of calculated and simulated storm surge at two station for 3 cases of wind + pressure, wind only and pressure only (data from WRF model). The contribution of wind is mainly.

-1

-0.5

0

0.5

1

1.5

9/15/2014 12:00 9/16/2014 12:00 9/17/2014 12:00Stor

m s

urge

(m

)

Time (h)

OBS

Wind and pressure

Win only

Pressure only

Effect of tide storm surge

Comparison of simulated and observation storm surge. The simulated results for two case, with and without consider tide.

Effect of wave on storm surge

Comparison of simulated and observation storm surge. The simulated results for two case, with and without consider wave.

Future works:

- Meteorological tsunami (anormalyze water level) on the middle coast of Vietnam

- Wave and ocean current using radar data (assimilation)

local/rear/tail winds

front winds