DEVELOPING PROBABILISTIC SEISMIC HAZARD MAPS OF SAGAING, SAGAING

REGION, MYANMAR

December, 2015

CONTENTS EXECUTIVE SUMMARY ....................................................................................................... 3

1. INTRODUCTION ............................................................................................................ 5

1.1 Objectives of the project ............................................................................................. 6

1.2 Structure of report ........................................................................................................... 7

2 SEISMOTECTONICS AND GEOLOGY .............................................................................. 9

2.1 Seismotectonics of the region .......................................................................................... 9

3 METHODOLOGY AND USED DATA ............................................................................... 14

3.1 Methodology of Seismic Hazard Assessment ................................................................ 14

3.2 Applied Data ............................................................................................................. 15

3,2,1 Seismic Sources Identification and Characterization ............................................... 15

3.2.2 Site Investigation .................................................................................................... 16

3.3 Regional Geological Setting ...................................................................................... 18

3.4 Ground Motion Prediction Equations (GMPEs) .............................................................. 20

4 RESULTS ........................................................................................................................ 21

4.1 Site Condition ........................................................................................................... 21

4.2 Seismic Hazard ........................................................................................................ 25

4.2.1 Seismic hazards for 475 years recurrence interval .................................................. 26

4.2.2 Seismic hazards for 2475 years recurrence interval ................................................ 31

Bibliography ........................................................................................................................ 38

APPENDICES ..................................................................................................................... 41

Appendix A .......................................................................................................................... 42

Appendix B .......................................................................................................................... 43

Appendix (C) ....................................................................................................................... 44

Appendix (E) ....................................................................................................................... 50

2

EXECUTIVE SUMMARY

Sagaing City is one of the cities that is passed through by the most active fault in Myanmar,

the right-lateral, strike-slip, Sagaing Fault. The city has experienced many large earthquakes

since 14th century. The earliest record of the historical earthquakes is the 1429 ear thquake

and t his earthquake caused several walls were dam aged i n Sagaing – Innwa region. The

historical ear thquakes h appened i n and round this r egion ar e t he ev ents o f 1429 , 1467 ,

1501, 1602, 1696, 1771, 1776 1830 and 1839.

Among those earthquakes, 1839 Innwa (Ava) earthquake is likely to be the largest event and

the magnitude is currently assumed as > 7.5. This earthquake struck on March 23 at 4:00

am. Due to this earthquake the city wall of Amarapura, and several buildings were collapsed.

The ground surface fractures were formed in the cities of Amarapura, Innwa and Sagaing,

together with the ground water poured out. According to the records, nearly all of the

houses, pagodas, and monasteries were completely damaged in Innwa. At least 300 to 400

casualties w ere r esulted in I nnwadue t o t his e vent. The g round surface f ractures were

formed al ong t he A yeyarwady R iver, bet ween I nnwa and A marapura and s ome ground

subsidence w ere al so occurred i n t he w idth o f 5 t o 20 feet. S everal af tershocks al so

happened i n t he nex t s ix m onths. T he most recent ev ent happened i n t his ar ea is t he

magnitude 7 .0, 1956 S againg E arthquake and i t s truck on J uly 16 at 9: 40 pm . This

earthquake c aused s everal pa godas, and bui ldings w ere da maged an d about 40 dea ths

happened in Sagaing.

According to the above mentioned earthquakes histories, Sagaing can be regarded

as one of the cities that have high probability of the large earthquake (≥ 7.0 magnitude)

potential in the f ugure. T herefore, Myanmar G eosciences S ociety ( MGS), M yanmar

Engineering Society (MES) and Myanmar Earthquake Committee (MEC) conducted the

3

seismic risk as sessment for S againg C ity ( Sagaing R egion)in 2013 , w ith t he ai d of t he

United Nations Human Settlements Programme (UN-HABITAT). This project includes two

parts: the seismic hazard assessment (SHA) and seismic risk assessment (SRA), MGS and

MEC conducted SHA, while MES performed SRA. This report is for SHA for Sagaing City,

Sagaing Region.

In dev eloping the s eismic haz ard maps for S againg, pr obabilistic s eismic ha zard

assessment (PSHA) method is used. We developed the seismic hazard maps – peak ground

acceleraiton (PGA) map, spectral acceleration (SA) maps for the natural periods of 0.2 s, 0.3

s and 1.0 s and peak ground velocity (PGV) maps - for 10% probability of exceedance in 50

years (475 years return period) and 2 % probability in 50 years (2475 years return period).

These hazard maps are hopefully used for the purposes to mitigate the effects of the

earthquakes, es pecially in des igninng for t he seismic s afety for the current and future

constuction o f varioius sorts o f bui ldings, planning o f the retrifiting o f t he ex isting bui lding,

land-use planning of the city, and all of the preparedness schemes for the city.

4

1. INTRODUCTION

Sagaing is the previous capital of Sagaing region, Myanmar, located in the southern end of

the region. The population of the city is about 100,000 according to 2014 census. According

the historical earthquake records, the city has experiences several large magnitude

earthquake as l isted i n Table ( 1). The events happened i n Sagaing – Innwa ar ea ar e t he

earthquakes happened i n 1429, 1467, 1485, 15 01, 1620, 1646, 1648, 1660, 1690, 1696,

1714, 1771, 1776, 1830 and 1839. The most recent earthquake and the most affected one is

the Sagaing earthquake that struck on July 16, 1956. This magnitude 7.0 earthquake caused

several pagodas, and buildings were damaged and about 40 deaths happened in Sagaing.

However, the most deadliest event happened in this area is 1839 Innwa (Ava) earthquake

that struck on March 23 at 4:00 am. This earthquake is likely to be the largest event and the

estimated magnitude is > 7.5 (~8.0). This earthquake caused several buildings including the

city wall of Amarapura were collapsed. According to the records, near ly al l of t he houses,

pagodas, and monasteries were completely damaged in Innwa. At least 300 to 400

casualties were resulted in Innwa due to this event. The fractures were formed in the cities

of Amarapura, Innwa and Sagaing, especially along the Ayeyarwady River between Innwa

and Amarapura, together with the ground water poured out. Some ground subsidence were

also occurred in the width of 5 to 20 feet. Several aftershocks also happened in the next six

months.The other event not happened near Sagaing area, but happened within the 250 km

radius, is May 23, 1912 Maymyo earthquake.

The main seismogenic active fault that can contribute the major seismic hazard is the right-

lateral, strike-slip Sagaing Fault that is passing through the city. Kyaukkyan Fault, Nampon

Fault, Shweli Fault, Moemeik Fault, West BagoYoma Fault and Gwegyo Fault are the other

seismogenic s ources for S againg.Therefore, t he c ity c an ex pect the l arge ear thquake t o

happen in the future and its effects can be large for the city.

5

In recent year 2012, the magnitude 6.8 Thabeikkyin Earthquake happened on November 11

in the north of Sagain at about 125 km. Various kinds of buildings such as pagodas, houses

and schools, about 500 were destroyed by this event, resulting 26 de aths and 231 injuries.

This earthquake also affected Sagaing City and one building damaged and walls of several

pagodas were fractured.

On the other hand, the urban development of Sagaing is also increasing along the

Sagaing Faul t. M oreover, new pr ojects of i nfrastructures c onstruction and bui lding

construction are c ontinuing. Therefore M EC, MGS and M ES i mplemented t o dev elop t he

seismic hazard maps and risk maps of Sagaing, with the aids of the United Nations

Development Programs (UNDP).

1.1 Objectives of the project

The main goal of the project is to construct the seismic hazard maps and seismic risk maps

of Sagaing, Sagaing Region. The objectives of the project of seismic hazard assessment of

Sagaing include the following:

1. To develop the probabilistic seismic maps of the city, the seismic hazard maps will

show t he hazard par ameters o f peak ground acceleration (PGA); s pectral

acceleration (SA) at the periods of 0.2 s, 0.3 s and 1.0 s; and pea k ground velocity

(PGV). These seismic hazard maps will correspond to 10% probability of exceedance

in 50 y ears ( 475 y ears return per iod) and 2% probability i n 50 years ( 2475 y ears

return period).

2. To contribute these seismic hazard assessment results to the corresponding

organizations that will include the civil societies, the ministries and depar tments that

will have to use for seismic safety des igns development, retrofitting for the seismic

unsafe buildings and land-use planning, etc.

6

3. To pr ovide t he r esults t o t he r espective depar tments and or ganizations ( probably

publics) for earthquake disaster education and preparedness purposes.

By s eeing t he abov e m entioned obj ects, the mitigation o f ea rthquake effects o n t he

peoples of Sagaing, and build-in environment is the major purpose of this project.

1.2 Structure of report

The r eport i s composed of five chapters and t he chapter 1 i ntroduces the s ituation o f t he

seismicity and t ectonics s ituation with respect to t he current s ituation o f the c ity, Sagaing,

together with objectives of the project. The chapter 2 discusses the seismicity of the city and

its region, correlating with the regional tectonics, and the geology of the area, since the

surface geology is one of the important parameters that strongly influence on the earthquake

damage p roperties. The m ethodology and research p rocedure appl ied i n t his p roject work

comprise o f t he chapter 3. The dat a appl ied i n the seismic hazard assessment works a re

discussed i n t his c hapter t o under stand the a dvantages o f the us age and i ts l imitation.

Chapter 5 presents the results of seismic hazard assessment, and the seismic hazard maps

of Sagaing for 10% and 2% probabilities o f exceedance in 50 y ears (475 years and 247 5

years return periods). The PGA maps, SA (0.2 s, 0.3 s and 1.0 s) maps, and PGV maps are

the main outputs of the project and the average shear wave velocity to the upper 30 m (Vs30)

contour map is also included. As a final chapter of the report, the discussion on the results of

the project and the recommendation for the earthquake disaster mitigation for Sagaing are

presented in chapter 5.

7

Figure (1) Map of the location of the project Sagaing City.

8

2 SEISMOTECTONICS AND GEOLOGY

2.1 Seismotectonics of the region

When the seismicity of Myanmar is observed as the whole country, most of the crustal faults

such as the major right-lateral s trike-slip faults o f Sagaing Faul t, Kyaukkyan Faul t (KK F .)

and Nampon Fault (NP F.); the left-lateral strike-slip faults in Shan-Tanintharyi Block such as

Moemeik Faul t, S hweli Faul t; and t hrust s ystems o f West B agoYoma Faul t, and

Gwegyomostly generate the shallow focus earthquakes (≥ 40 km in focal depth).

Among them, the Sagaing Fault is the major active fault, running through or near the

major cities such as Yangon, Bago, Taungoo, Naypyitaw, Pyinmana, Meikhtila, Sagaing,

Mandalay, Wuntho and Myitkyina. The length of the fault is above 1200 km as the total, and

the s lip r ate i s from 18 – 22 mm/yr (Wang Y u et al ., 2013) . The m ajor ev ents ( M > 7. 3)

generated by t his f ault ar e t he w ell-known 1 839 Ava ( Innwa) earthquake, 1929 S wa

earthquake, 1930 B ago ear thquake, 1930 P hyu ear thquake, 1931 H tawgaw ear thquake,

1946 two continuous Tagaung earthquakes, and 1956 Sagaing earthquake. The slip rate of

West BagoYoma Fault is 5 m m/yr, as the largest rate, while that of other faults is around 1

mm/yr (SoeThuraTun et al., 2011).

Rather than the seismicity related to the crustal faults, the other seismogenic sources are the

subduction z one of I ndian P late beneat h B urma P late i n t he west of Myanmar and t he

collision zone of Indian Plate with Eurasia Plate in the northwest. While the rate of collision is

about 50 m m/yr, t he s ubducted r ate i s 36 m m/yr(Socquet et al ., 2006 ). Other t ectonically

seismogenic source Adaman spreading region. The spreading rate is about 37 mm/yr and

the s eismicity happened i n t his r egion mostly c omprises the s hallow f ocus ev ents. 1762

Arakan ear thquake i s pr obably t he subduction r elated event and t he m agnitude i s around

7.5M. From t he c ollision z one of I ndian and E urasia P lates, t he l argest ev ent i s t he

magnitude 8.6, August 8, 1950 earthquake.

9

The seismicity of Myanmar is depicted in Figure (2) and Figure (3) i llustrates the seismicity

of Sagaing area. Figure ( 4) p resents the magnitude > 7.0 earthquakes happened in and

around the city. Table (1) lists the previous historical and instrumental recorded significant

events, describing the respected properties of damages and casualties.

Figure (2) Seismicity map of Myanmar (ISC earthquake catalog, 1906 – 2011)

10

Figure (3) Seismicity map of Sagaing area.

11

Figure (4) Map of the previous magnitude ≥ 7.0 events happened around Sagaing

area

12

Table (1) List of the previous earthquakes happened in and around Sagaing Date Location Magnitude or brief description

1429 Innwa Fire-stoping enclosure walls fell

1467 Innwa Pagodas, solid and hollow, and brick monasteries destroyed

24, July, 1485 Sagaing 3 well-known pagodas fell

1501 Innwa Pagodas, etc. fell

6, June, 1620 Innwa Ground surface broken, river fishes were k illed af ter quake

10, Sept, 1646 Innwa

11, June, 1648 Innwa

1, Sept, 1660 Innwa

3, Apr, 1690 Innwa

15, Sept, 1696 Innwa

8, Aug, 1714 Innwa 4 well-known pagodas destroyed

15, Jul, 1771 Innwa

9, June, 1776 Innwa A well-known pagoda fell

26, April, 1830 Innwa

21, Mar, 1839 Innwa Old palace and many buildings demolished; pagodas and city walls fell; ground surface broken; the river’s flow w as r eversed f or sometime; M ingun P agoda shattered

23, May, 1912 Taunggyi M = 8. 0, al most al l of c ities i n Myanmar w ere shocked

16, July, 1956 Sagaing Several pag odas and b uildings severely dam aged, about 40 deaths

13

3 METHODOLOGY AND USED DATA

3.1 Methodology of Seismic Hazard Assessment

In conducting seismic hazard assessment for Sagaing, pr obabilistic methodology is used

and it includes four steps (Cornell, 1968, McGuire, 1976, Reiter, 1990 and Kramer, 1996).

The following the basic steps of probabilistic seismic hazard assessment (PSHA):

1. Identification of seismic sources: the seismic sources such as the fault sources, areal

or volumetric sources from t hose the ear thquake pot entials o f l arge magnitude can be

expected to happen in the future and can generate the significant ground motion at the

city are identified in this stage.

2. Characterization of seismic sources: the seismic source parameters for each identified

seismic s ources (fault, ar eal or v olumetric s eismic source) ar e calculated and the

parameters es timated ar e t he s patial and temporal oc currence pa rameters s uch as a-

and b- values, the annual recurrence of the earthquake of the certain magnitude, and the

maximum earthquake potential. For fault seismic sources, the fault parameters such as

the its geometry and geological parameters such as the dip, fault length, slip rate, etc.

are also needed to estimate.

3. Choosing the ground motion prediction equation (GMPE): the predictive ground

motion equations are commonly applied in PSHA. By them, the ground motion at a s ite,

that c an be gener ated by an y pos sible s ized ear thquake ar e es timated. The m ost

suitable GMPEs are need to choose for the city based on the tectonic environments and

fault types, etc.

4. Integration of va riables t o es timate the se ismic ha zard: the s eismic haz ards, i .e.

PGA, SA (at the periods of 0.2 s, 0.3 s and 1.0 s) and PGV are estimated by considering

the unc ertainties o f t he location, t he m agnitude o f t he ear thquake, and ground m otion

parameters, with the combination of the effects of all the earthquakes with the different

14

magnitude from t he l ower bound m agnitude, di fferent di stance and di verse occurrence

probability.

In P SHA, t he t hree input par ameters: 1 ) s eismic s ources da ta t hat include t he f uture

earthquakes related parameters such as the maximum earthquake magnitude, the (temporal

and spatial) occurrences of the earthquakes with certain magnitude, etc., 2) the parameters

and coefficients o f the chosen G MPE, and 3) t he par ameters o f s ite condition, m ostly t he

average shear wave velocity to the upper 30 m (Vs30).

3.2 Applied Data

3,2,1Seismic Sources Identification and Characterization

In 2011, Myanmar Earthquake Committee (MEC) carried out the probabilistic seismic hazard

assessment for Myanmar and developed the PSHA maps of the country. In that assessment,

the s eismic s ources i dentification and characterization of the ac tive faults w as done by

SoeThuraTun et al. (2011) and they constructed the active fault database for Myanmar. In

the same work, Myo Thant et al. (2012) conducted the areal seismic sources identification

and characterization f or each t ectonic domain such as t he subduction zone o f I ndia P late

beneath B urma P late, i n t he w est of t he c ountry; t he c ollision z one o f India P late w ith

Eurasia Plate in the north and nor thwest, and the Andaman spreading center in the south.

While S oeThuraTun et al. ( 2011) c onstructed t he ac tive f aults dat abase, t he geological

information and paleoseismologic data such as the geometry of the fault, dip and strike of

the fault, fault displacement, fault s lip (slip per event or annual s lip rate), etc. are appl ied,

Myo T hant e t al . (2012) appl ied t he s eismological and geological i nformation s uch as

instrumental ( ISC ear thquake c atalog, 1900 – 2011; A NSS c atalog 19 36 – 2011) and

historical records of the previous events and t he geological parameters such as the rate of

subduction, collision, and spreading, and the age of subducted slab, etc.

15

For the present seismic hazard assessment, from the seismic sources identified by

SoeThuraTun et al. (2011) and Myo Thant et al. (2012), those lie within 250 km in radius are

obtained as t he m ost p ossible s eismic s ources ( fault and ar eal) t hat c an c ontribute t he

seismic hazards to Sagaing.

3.2.2 Site Investigation

The s ite geology dat a pl ays an i mportant r ole f or t he s ite s pecific s eismic haz ard m ap

development. In the site investigation of Sagaing, the borehole drilling is performed in four

locations in the city with reference to the geology and geomorphology. The SPT test and soil

sampling a re c arried o ut w hile t he bor eholes ar e dr illing. Labo ratory t ests o f s ome s oil

samples are also conducted in this project.

As t he o ther s ite i nvestigation m ethod, we conducted the microtremor surveying in

Sagaingduring 7 – 11 July, 2013 as geophysical s urvey. T he s ite par ameter i n seismic

hazard c alculation by us ing the selected G MPEis t he av erage s hear wave v elocity t o t he

upper 30 m, Vs30 and the H/V spectral technique is used to calculate Vs

30.

The l ocations o f bor eholes and m icrotremor s urvey s ites i n S againg are s hown i n

Figure (5).

16

Figure (5) Map depicting the locations of boreholes and microtremor survey points

17

3.3 Regional Geological Setting

It can be generally said that Sagaing City is bounded by Minwun range in the west and

Sagaing Hill in the east. While Minwun Range is covered by Minwun Metamorphics

composed mainly of m etapelite and m etabasite, S againg H ill i s oc cupied by S againg

Metamorphics that are mainly gneisses and Calciphyres with white marble bands

(MyintThein, 2009) . R ather t han M inwun Metamorphics, I rrawaddy Fo rmation and U pper

Pegu Group can also be seen in Minwun Range.

Most parts of Sagaing are covered by alluvium, and river terrace, especially in the southern,

central, western and eas tern parts of the city. The north-western part of the city is covered

by Minwun Metamorphics and north-eastern part by Sagaing Metamorphics. Regional

geological map of the Sagaing City is shown in Figure (6).

The c ity i s m oreover pas sing t hrough(the c enter of the c ity) by right-lateral, s trike-slip

Sagaing fault;and bounded by Kyaukkyan Fault and Nampon Faul t, and West BagoYoma

Fault and Gwegyo Faultin the west.

18

Figure (6) Regional geological map of Sagaing

19

3.4 Ground Motion Prediction Equations (GMPEs)

After t he g round motion v alues ( peak ground ac celeration (PGA), s pectral

acceleration (SA) at the periods of 0.2 s, 0.3 s and 1.0 s, and peak ground velocity

(PGV)) calculated by using the several different ground motion prediction equations

(GMPEs) are correlated, the GMPE of Boore et al. (1997) is used for seismic hazard

calculation of PGA and SA, and Boore and Atkinson (2008) NGA is applied for PGV

calculation.

20

4 RESULTS

4.1 Site Condition

From borehole drilling, SPT and Labor atory analysis, the N-values and dens ity of each soil

layer ar e obt ained. T hese par ameters a re t he bas ic par ameters for H /V s pectral r atio

analysis f or mircrotremor dat a. A s i n al l t he g eophysical m ethods, the ac tual g eological

condition c an be appl ied as t he m odel f or m icrotremor dat a anal ysis. T he s hear w ave

velocity s tructure o f eac h s urvey s ite i s c onstructed bas ed on t his m odel and f inally we

estimate the average shear wave velocity to the upper 30 m, Vs30. For example, Figure (8)

shows t he H /V spectral r atio o f t he m icrotremor s ite SG-30 and Figure (9) r epresents t he

shear wave velocity structure. The average shear wave velocity to the upper 30 m, Vs30 of

this site is obtained as 195.64 m/s.

The o ther ex ample i s t he s hear w ave v elocity o f s tructure der ived from t he H /V

spectral ratio analysis of the microtremor measurement at the site SG-13. While Figure (10)

shows the H/V spectral ratio of the site SG-13, Figure (11) illustrate the shear wave velocity

structure of the site. The average shear wave velocity to the upper 30 m, Vs30 is deduced as

382.69 m/s.

From about H/V spectral ration analysis of about 39microtremor survey sites, the

shear wave velocity structure of all sites are developed and Vs30 of all sites are estimated.

The Vs30 contour map of Sagaing city is f inally developed by interpolating these Vs

30 values

of 39 sites. Figure (12) shows the Vs30 contour map of Sagaing and this map is applied as

the parameter o f s ite condition f or t he seismic hazard calculation.Most of the soil t ypes in

Sagaing can be classified as very dense soil (i.e. B class) based on the Vs30 values (Uniform

Building Code, UBC and Eruocode 8, EC8). Soft soil (D) can only be obs erved in southern

and western margin of the city. As mentioned in previous session of regional geology of the

21

city, t he nor th-western and nor th-eastern par t o f t he c ity is covered by hard r ock, i .e. s ite

class A.

The V s30 value i s t he m ain i nput par ameter o f t he site c ondition, for t he s eismic

hazard (peak ground acceleration (PGA); spectral acceleration (SA) at the periods of 0.2 s,

0.3 s and 1. 0 s ; and pe ak ground velocity ( PGV)) calculation by us ing t he gr ound m otion

prediction equation (GMPE).

Figure (8) H/V spectral ratio of the microtremor survey site, SG-30

0

1

2

3

0.1 1 10

H/V

Spe

ctra

l rat

io

Frequency [Hz]

observed data

initial model

22

Figure (9) Shear wave velocity structure of the microtremor survey site, SG-30 (Vs30-195.64 m/s)

Figure (10) H/V spectral ratio of the microtremor survey site, SG-13

-70

-45

-20

0 100 200 300 400

Dep

th (m

)

Vs (m/s)

0

1

2

3

4

0.1 1 10

H/V

Spe

ctra

l rat

io

Frequency [Hz]

observed data

initial model

23

Figure (11) Shear wave v elocity s tructure of the microtremor survey s ite, SG-13 (Vs30 -382.69 m/s)

-200

-175

-150

-125

-100

-75

-50

-25

0

0 300 600 900 1200

Dep

th (m

)

Vs (m/s)

24

Figure (12) The Vs

30 contour map of Sagaing

4.2 Seismic Hazard

The seismic hazard assessment is carried out for 10% and 2% probabilities of exceedance

in 50 years (475 years and 2475 years recurrence intervals) by using PSHA. The results of

seismic hazard assessment will be discussed in this session.

25

4.2.1 Seismic hazards for 475 years recurrence interval

The s eismic haz ard p resented by m eans o f pe ak ground ac celeration (PGA) i n g for t he

recurrence interval (10% probability of exceedance in 50 years) is shown in Figure (13). The

maximum seismic hazard area is in the west of the city and the PGA ranges from 0.65 g to >

1.0 g. The eastern part of Zayar Ward, the western part of Thaw Tar Pan are in the highest

seismic z one, P GA > 1 .0 g. H owever, m ost par ts o f t he S againg s uch as t he wards o f

Takaung, Poe Tan, More Zar, Aye Mya Wati, Nanda Wun, PannPae Tan, Seingone, Myothit,

Htone B o, P arami, Y warHtaung, Zay ar ( western par t), Thaw T ar P an ( eastern par t),

MeeYahtar, Shweminwun, and Patamyar comprise the second most highest seismic hazard

zone (PGA – 0.9 g – 1.0 g).

From Figure (14) to (16) depict the probabilistic seismic hazard maps presented in

terms of spectral acceleration at the periods of 0.2 s, 0.3 s, and 1.0 s, for 10% probability of

exceedance in 50 years (475 years recurrence interval). The range of SA is from 1.4 g to 2.2

g for 0. 2 s pe riod and 0 .5 g to 1 .8 g for 0. 3 s period. H owever, S A at the per iod o f 1. 0s

ranges f rom < 0. 6 g t o >1.4 g. The S A v alues of these nat ural per iods f or 475 y ears

recurrence interval can be used for developing the seismic safety design for the (ordinary)

buildings and structures.

26

Figure (13) P robabilistic s eismic h azard m ap o f S againg, S againg Region, f or 10% probability of exceedance in 50 years (475 years recurrence interval), in terms of peak ground acceleration (PGA) in g.

27

Figure ( 14) Probabilistic s eismic haz ard m ap o f S againg c ity, S againg Region f or 10% probability of exceedance in 50 years (475 years recurrence interval), in terms of spectral acceleration (SA) at the period of 0.2 s, in g.

28

Figure ( 15) P robabilistic s eismic haz ard m ap o f S againg c ity, S agiang Region f or 10% probability of exceedance in 50 years (475 years recurrence interval), in terms of spectral acceleration (SA) at the period of 0.3 s, in g.

29

Figure ( 16) P robabilistic s eismic haz ard m ap o f S againg city, S againg Region f or 10% probability of exceedance in 50 years (475 years recurrence interval), in terms of spectral acceleration (SA) at the period of 1.0 s, in g.

30

4.2.2 Seismic hazards for 2475 years recurrence interval

Figure (17) shows the probabilistic PGA map of Sagaing for 2% probability of exceedance in

50 years (2475 years recurrence interval). The seismic hazard zones are trending in NW-SE

and the PGA values range from 0.8 g to > 1.5 g. The highest PGA is contributed to some

portions of the city with the ground motion level of > 1.5 g and it includes Htone Bo, Parami,

Thaw Tar Pan, Zayar, MeeYahtar, Shweminwun and Patamyar Wards. Some parts of wards

such as Aye Mya Wati, Poe Tan also fall in this seismic hazard zone. The rest wards of the

city – Takaung, Seingone, Myothit, Moe Zar, Nanda Wun, Aye Mya Wati, DarwaeZay,

PannPae Tan, YwarHtaung, Tat Myae and Nilar – belong to the second highest seismic

hazard zone and t he PGA of this zone is f rom 1.35 g to 1.5 g. For 2,475 years recurrence

interval, the city falls in these two zones.

Probabilistic SA maps for the periods of 0.2 s, 0.3 s, and 1.0 s are illustrated in Figures (18-

20). The hazard distribution patterns of these ground motion parameters are also nearly the

same w ith each ot her. To dev elop t he s eismic r esistance des ign for l ong t erm projects

(buildings and structures), t he S A for 2475 y ears r ecurrence i nterval c an be us ed for t his

city.

31

Figure (17) Probabilistic seismic hazard map of Sagaing, Sagaing Region, for 2% probability of exceedance in 50 years (2475 years recurrence interval), in terms of peak ground acceleration (PGA) in g.

32

Figure ( 18) P robabilistic s eismic haz ard m ap o f Sagaing c ity, S againg Region f or 2% probability of exceedance in 50 years (2475 years recurrence interval), in terms of spectral acceleration (SA) at the period of 0.2 s, in g.

33

Figure ( 19) P robabilistic s eismic haz ard m ap o f Sagaing c ity, S againg Region f or 2% probability of exceedance in 50 years (2475 years recurrence interval), in terms of spectral acceleration (SA) at the period of 0.3 s, in g.

34

Figure ( 20) P robabilistic s eismic haz ard m ap o f Sagaing c ity, S againg Region f or 2% probability of exceedance in 50 years (2475 years recurrence interval), in terms of spectral acceleration (SA) at the period of 1.0 s, in g.

35

5 DISCUSSION AND RECOMMENDATION

With t he ai ds o f U NHABITAT, Myanmar E arthquake C ommittee (MEC), Myanmar

Geosciences Society (MGS) and Myanmar Engineering Society (MES) carry out the seismic

hazard and r isk assessment for three cities: Sagaing City (Sagaing Region), and B ago and

Taungoo Cities (Bago Region). This report is prepared for the probabilistic seismic hazard

maps of Sagaing. While MES develops the seismic risk maps of Sagaing, MEC and MGS

develop the seismic hazard maps of the city by using the probabilistic seismic hazard

assessment m ethodology ( PSHA). We c onstruct t he s eismic haz ard m aps of S againg for

475 years recurrence interval (10% probability of exceedance in 50 years) and 2475 y ears

recurrence interval (2% probability of exceedance i n 50 years. The hazard maps include

probabilistic peak ground acceleration (PGA) map, and spectral acceleration (SA) maps of

0.2 s, 0.3 s and 1.0 s. There are, therefore, four seismic hazard maps for each recurrence

interval level.

Instead o f al l s eismic haz ard m aps, t he di scussion will m ainly c oncern t o P GA gr ound

motion l evel of t he c ity, f or bot h 475 y ears and 2475 years r ecurrence i ntervals. For 475

years r ecurrence i nterval, t he P GA l evel of the c ity i s in t he r ange o f 0 .6 g t o > 1. 0 g.

Therefore, it can be regarded as the city is in violent zone of perceived shaking, heavy zone

of potential damage, and instrumental intensity zone IX. However, for 2475 years recurrence

interval, the PGA level of most parts of the city is in the range of 0.8 g to > 1.5 g. It means

that the city comprises the extreme zone of perceived shaking, very heavy zone of potential

damage, and X+ zone of instrumental intensity.

The s eismogenic s ource t hat can m ainly c ontribute t he seismic haz ard t o S againg is t he

right-lateral, s trike-slip S againg Faul t which i s pas sing through the c ity. T herefore, the

seismic hazard level is very highfor the center of city, trending NS direction. This is the key

point to be considered for land-use planning or urban development.

36

We develop the seismic hazard m aps o f S againg city by us ing t he c urrent available dat a

such as the seismic sources data, and s ite data, etc. However, i t is needed to update and

modify these maps based on the availability of more data, especially on the seismic sources

data such as the active fault data, paleoseismological data, etc. These maps can be used for

developing the seismic resistance designs of buildings, structures, infrastructures for certain

projects. But it should be noted that it is need to adjust what hazard maps, i.e. the different

recurrence interval level (either 475 years recurrence level or 2475 years recurrence level)

should be us ed bas ed on t he pr ojects pu rposes. H owever, al l m aps c an be used for

earthquake disaster preparedness purposes.

For special or major project, it might be needed t o conduct the site specific seismic hazard

analysis for that site or location.

37

Bibliography

Aki, K . 1965 . M aximum l ikelihood es timate o f b i n t he formula l og N = a-bm and i ts

confidence limits, Bull. Earthq. Res. Inst., Univ. Tokyo, 43, 237-239.

Ambraseys, N . N . 1988 . Magnitude – Fault Lengt h R elationships f or Earthquakes i n t he

Middle East, In: Lee, W.H., Meyers, H. &Shimazaki, K. eds, Historical Seismograms

and Earthquakes of the World, Acad. Press Inc., 309-310.

Atkinson, G. M . 1984. Attenuation o f S trong Ground M otion i n C anada from a R andom

Vibrations Approach, Bulletin of the Seismological Society of America, Vol. 74. No.

6, pp. 2629-2653

Bender, F., 1983, Geology of Burma. 225p.

Boore, D .M., J oyner, W.B., and Fu mal, T.E. 1 997. E quations for E stimating H orizontal

Response S pectra an d P eak A cceleration from Western N orth A merican

Earthquakes: A Summary of recent Work, Seismological Research Letters, Vol. 68,

No. 1, 128-153. (http://iisee.kenken.go.jp/ eqflow/reference/ 1 12.htm

Cornell, C. A. 1968. Engineering Seismic Risk Analysis, Bulletin of the Seismological Society

of America, Vol. 58, 1583-1606.

Gutenberg, B., and R ichter, C. F. 1944. Frequency of Earthquakes in California, Bulletin of

the Seismological Society of America, 34: pp. 185-188.

International S eismological C entre (ISC), 201 1, On-line B ulletin, I nternat.Seis. Cent.,

Thatcham, United Kingdom, http://www.isc.ac.uk/Bull.

KhinThetSwe and Myo Thant, 2012. Probabilistic Seismic H azard Maps of B ago R egion,

Myanmar, 1 st International C onference on R egional G eology, S tratigraphy and

Tectonics o f M yanmar and N eighboring C ountries and E conomic G eology

(Petroleum and Mineral Resources) of Myanmar

38

KhinThetSwe, 2012. Seismic H azard A ssessment of B ago R egion by using P robabilistic

Seismic Hazard Analysis (PSHA). Department of Geology, Yangon University. 16-18

p.

Kijko, A. 2004. Estimation of the Maximum Earthquake Magnitude, mmax, Pure and A pplied

Geophysics, Vol.161, No.8. pp. 1655-1681.

McGuire, R. K. 1976. Fortran computer program for seismic r isk analysis, US. Geol. Surv.,

Open - File Rept 76-67, 90 pp.

MaungThein and T int Lw inSwe, 2005. The S eismic Zone M ap of M yanmar, Myanmar

Earthquake Committee, Myanmar Engineer Society.

Myo T hant, 2010 . Lecture N otes o f E arthquake E ngineering ( Part-1). Department o f

Engineering Geology, Yangon University. 32 p.

Myo T hant, Nwe Le′ Nge, SoeThuraTun, MaungThein,WinSweand ThanMyint, 2012.

Seismic H azard A ssessment M yanmar, Myanmar E arthquake C ommittee(MES),

Myanmar Geosciences Society(MGS).

Nwe Le′ Nge, 2010. Evaluation of Strong Ground Motion for the Central Portion ofYangon.

Department of Geology, Yangon University. 62 p.

Papazachos B. C., Scordilis E. M., Panagiotopoulos, D. G., Papazachos, C. B. and

Karakaisis G . F . 2004 . Global R elations be tween S eismic Faul t P arameters an d

Moment M agnitude o f E arthquakes, Proced. of 10 th International C ongress,

Thessaloniki, April, pp. 1482-1489. (in Appendix A)

Reiter, L 1990. Earthquake H azard A nalysis- Issues and I nsights, C olumbia University

Press, New York, 254pp.

San Shwe&MaungThein, 2011. Seismic Microzones of Bago-OaktharMyothit Area, Journal

of the Myanmar Geoscience Society, 66 p.

39

Steven L. Kramer, 199 6. Geotechnical E arthquake E ngineering, Civil E ngineering and

Engineering Mechanics, University of Washington. 19-20, 45-50 p,595p.

United Nations, 1996. Geology and M ineral Resources of Myanmar. Economic and S ocial

Commission for Asia and the Pacific. 183 p.

Win Swe and Win Naing, 2008. Seismicity and Major Active Faults of Myanmar, Myanmar

Geoscience Society, Yangon, Myanmar.

40

APPENDICES

41

Appendix A The maximum magnitude of earthquake potential expected to happen by fault specific

sources c an be det ermined by us ing t he following relationships of earthquake m agnitude

and fault length.

Inoue et al., AIJ (1993); 0.5M = Log L + 1.9 (A-1)

Ambraseys’s equation (1988); Msc= 1.43 logL + 4.63 (A-2)

in which Msc is the expected surface wave magnitude and L is the fault length.

Mohammadioun&Serva (2001); Ms= 2 log L + 1.33 log ∆σ + 1.66 (A-3)

where, Msis t he surface wave m agnitude, L is t he f ault r upture l ength ( km) and ∆σ is t he

stress drop released by the earthquake (in bars) that depends on the width of the faults and

type ( kinematics) o f t he faults. S tress d rop pa rameters for eac h fault are c alculated by

applying the following relationships (Mohammadioun and Serva, 2001);

∆σN = 10.6 x W0.5 (A-4)

∆σSS = 8.9 x W0.8 (A-5)

∆σR = 4.8 x W1.6 (A-6)

in w hich ∆σN,∆σSS and ∆σRare stress drop (in bars) for normal, strike-slip and reverse

faults and W is the fault width (km) which is also determined by utilizing the relation of fault

length and fault width; L = 2W (Bormann and Baumbach, 2000).

M = (LogL+6.4)/1.13 (Ambraseys and Zatopek, 1968) (A-7)

M = 2.0 logLmax+ 3.6 (Otsuka, 1964) (A-8)

M = 2.0 logLmax+ 3.5 (Iida, 1965) (A-9)

M = 2.0 logLmax + 3.7 (Yonekura, 1972) (A-10)

in which L maxis the maximum earthquake fault length,

42

M = 1.7 LogL + 4.8 (Matsuda, 1977) (A-11)

and, 0.5 M = Log L + 1.86 for oblique faults (A-12)

0.59 M = Log L + 2.3 for Strike slip faults (A-13)

(Papazachos et al., 2004)

Appendix B

The maximum magnitude of the earthquake potentials which can be originated from all areal

seismic sources are determined by using the relationship of Kijko (2004);

)exp()}]exp(/{)}()([{ min22211maxmax nmnnEnEmm obs −+−−+= β (B-1)

where, E1(z) = {(z2 + a1z + a2)/ [z (z2 + b1z + b2)]} exp (-z) (B-2)

n1 = n / {1 - exp [-β(mmax - mmin)]} (B-3)

n2= n1exp [-β(mmax- mmin)] (B-4)

in which n is the number of earthquakes greater than or equal mmin, a1 = 2.334733,

a2 = 0.250621, b1 = 3.330657, and b2 = 1.681534.

It must be noted that Equation 2.23 does not constitute a direct estimator formmax since

expressions n1and n2, which appear on the right-hand side of the equation, also contain

mmax. Generally the assessment of mmax is obtained by the iterative solution of Equation (B-

1).

However, when mmax- mmin ≤ 2, and n ≥ 100, the parameter mmax in n1 and n2 can be

replaced by mmax(obs), pr oviding mmax estimator which can be obt ained w ithout iterations

(Kijko, 2004).

43

Appendix (C)

The mathematical expression of the probability of the ground motion parameter Z will exceed

a specified value z, during a specified time period T at a given site is as follow:

tzvezZP ⋅−−=> )(1)( (C.1)

wherev(z) is the mean annual rate of events from which the ground motion parameter Z will

exceed z at a certain site resulting from the earthquakes from all seismic sources in a region.

It can be calculated by applying the following equation:

∫ ∫∑ >⋅==

drdmrmzZPrfmfmzv RM

N

ni ),/()()()()(

1λ

(C.2)

where )( imλ = t he frequency of ea rthquakes on s eismic s ource nabove a m inimum

magnitude of engineering significance, mi ;

)(mfM = t he pr obability dens ity f unction o f ev ent s ize on s ource nbetween m0 and

maximum earthquake size for the source, mu;

)(rfR = t he pr obability dens ity f unction for di stance to ear thquake r upture o n s ource n,

which may be conditional on the earthquake size; and

P(Z>z|m,r)= the pr obability t hat, at a given a magnitude mearthquake and at a di stance

rfrom the site, the ground motion exceeds value z.

Therefore the calculation of the seismic hazards will be included the following steps;

1) Calculating the frequency of the occurrence of the event of magnitude m on source n,

2) Computing the probability density function of event size on source n

betweenm0 and mu,

3) Computing the probability distribution for the distance from the site to source n where the

event with the magnitude m will occur, and

44

4) Calculating, at each distance, the probability that an event with magnitude m will exceed

the specified ground motion level z, i.e. calculating the ground motion amplitude parameters

for a certain recurrence interval.

The seismic hazard values can be obtained for individual source (zones) and then combined

to express the aggregate hazard. The pr obability of exceeding a particular value Z, of a

ground m otion par ameter, z, is calculated f or on e pos sible ear thquake at one pos sible

source location and then multiplied by the probability that the particular magnitude

earthquake w ould oc cur at that par ticular l ocation. The p rocess i s t hen r epeated for al l

possible m agnitudes an d l ocations, and t hen summed al l o f t he pr obabilites on t hese

variables (Kramer, 1996).

Calculation of the Event Rate

The first step is the computation of the rate of occurrence of events of magnitude m.

The annual rate of exceedance for a pa rticular magnitude can also be determined by using

Gutenberg-Richter recurrence law.

Log Nc(m) = a – bm (C.3)

where Nc(m) is the yearly occurrence rate of earthquakes with magnitude ≥ m in a particular

source zone, a and b are constants specific to the seismic source zone, and t hese can be

estimated by a l east s quare anal ysis of the da ta bas e o f t he pas t s eismicity from eac h

seismic s ource. These values m ay v ary i n s pace and time. While the a-value g enerally

characterizes the level of seismicity in a given area i.e. the higher the a-value, the higher the

seismicity, the b-value describes the relative l ikelihood of large and s mall earthquakes, i .e.

the b-value increases, the number of larger magnitude earthquakes decreases compared to

smaller.

45

Probability of the Event Magnitude

The second step of the seismic hazard analysis is the calculation of the probability

that the magnitude will be within an interval of the lower bound magnitude ml and the upper

bound magnitude mu. It can be calculated by the following relation:

)()](exp[1

)](exp[)/()( 0max

0luul

M mmmm

mmmmmmPmf −−−−

−−=<<=

βββ

(C.4)

where, β = 2. 303b, mmax is t he m aximum magnitude o f t he ear thquake pot ential for a

specific seismic source (Kramer, 1996).

Probability of the Source-to-site Distance

The probability for the source-to-site distance can be computed as the same in the

second step and can be expressed by the following equation:

)()](exp[1

)](exp[)/()( 0max

0luul

R rrrr

rrrrrrPrf −⋅−−−

−−=<<=

βββ

(C.5)

in which rmax is t he l ongest source-to-site di stance, r0 is t he shortest distance, rl is t he

lower bound source-to-site distance, andru is the upper bound distance.

Probability of Ground Motion Parameter

The probability for a c ertain ground motion parameter, Z that will exceed z from the

specified magnitude, m and at the specific location (source) with the distance r, can be

calculated by utilizing the following relation:

))ln()ln((1),/(ln y

PHAzFrmzZPσ−

−=> (C.6)

46

where PHA is t he peak hor izontal ac celeration and σlny is t he s tandard dev iation of t hat

attenuation relation.By multiplying these probabilities from each sources and repeated again

for all possible seismic sources together with the above mentioned steps, the Probabilistic

PGA map can be developed for a certain area of interest or region.

Probability of Exceedance

The assumption called no memory (Poisson Model) is used the occurrence of certain

magnitude ear thquake i n any par ticular y ear, t he r eturn per iod ( T) of an ev ent

exceeding a par ticular ground m otion level i s r epresented by t he mathematical

expression as:

T = 1/v = - ∆t / ln (1 - P(Z>z)) (C.7)

In this equation, P(Z>z) is the desired probability of exceedance during the time T.

Appendix (D)

0

0.5

1

1.5

2

2.5

1906

1911

1916

1921

1926

1931

1936

1941

1946

1951

1956

1961

1966

1971

1976

1981

1986

1991

1996

2001

2006

2011

33.544.555.566.577.58

47

Figure (D-1). D iagram representing the relationship of normalized frequency of events of certain m agnitude w ith respect t o t ime ( year) for M yanmar r egion ( in w hich magnitude roundness is 0.25).

Table ( D-1). Time o f c ompleteness for t he ev ents w ith c ertain m agnitude for M yanmar region.

Magnitude Incremental Frequency

Time of

Completeness

3.0 791 1992

3.5 876 1992

4 1055 1978

4.5 541 1966

5 266 1964

5.5 89 1964

6 60 1933

6.5 33 1925

7 23 1925

7.5 6 1918

8 1 1906

8.5 1 1906

48

Figure (D-2). The Gutenberg-Richter relation for Myanmar region.

Figure (D -3). T he di agram i llustrating t he ann ual r ate o f ex ceedance o f c ertain magnitude earthquake for Myanmar region.

Log Nm = 4.744 - 0.8083m

-3.0

-2.0

-1.0

0.0

1.0

2.0

3.0

0 2 4 6 8 10

Log

Nm

Mw

0.001

0.01

0.1

1

10

4.0 5.0 6.0 7.0 8.0 9.0

Annu

al ra

te o

f exc

eeda

nce

Mw

49

Appendix (E)

Table (E-1) Ground profile (soil) types or classification of subsoil classes according

to UBC (Uniform Building Code) and EC8 (Eurocode 8) standards based ontheVs

30 values (modified from Sˆeco e P into 2002; Dobryet a l. 2000; Sabetta&Bommer 2002). (Source-Kanl1 et al., 2006).

Ground profile (Soil) type (UBC) or Subsoil Class (EC8)

Ground description (UBC)

Description of s tratigraphic pr ofile (EC8)

Shear wave velocity 𝑉𝑉𝑠𝑠30 (m s-1)

SA(UBC) Hard rock — >1500 (UBC)

SB(UBC) o r A (EC8)

Rock Rock or ot her rock-like geol ogical formation, including at most 5m of weaker material at the surface

760–1500 ( UBC) or >800

(EC8)

SC(UBC) o r B (EC8)

Very dense s oil and soft rock

Deposits of v ery de nse s and, gravel or v ery s tiff c lay, at l east several t ens of m i n thickness, characterized b y a gradual increase of m echanical pr operties with depth

360–760 ( UBC) or 36 0–800

(EC8)

SD(UBC) or C (EC8)

Stiff soil

Deep de posits of den se or medium-denses and, gravel or stiff clay with thickness from several tens to many hundreds of m.

180–360 (UBC and EC8)

SE(UBC) o r D (EC8)

Soft soil

Deposits of l oose-to-medium cohesionless s oil ( with or w ithout some s oft c ohesive layers), or of predominantly s oft-to-firm cohesive soil

<180 (UBC and EC8)

SF(UBC) or E (EC8)

Special soils

A s oil profile c onsisting of a surface —alluvium layer with V30s values of c lass C or D an d thickness varying between about 5 m and 20 m , under lain b y s tiffer material with

—

50

𝑉𝑉𝑠𝑠30 >800 m 𝑠𝑠−1

S1 (EC8) —

Deposits consisting—or containing a layer at least 10 m thick—of soft clays/silts with high plasticity index (PI >40) and high water content

<100 (EC8)

S2 (EC8) —

Deposits of l iquefiable s oils, of sensitive c lays, or an y ot her s oil profile not included in classes A–E or S1

— (EC8)

51

Photos of some pagodas affected by 2012 Thabeikkyin Earthquake

52

Photos of microtremor surveying

53

Microtremor instrument and its parts

54