VNU Joum al of Science, Earth Sciences 25 (2009) 10-19
A m e th o d to construct flood dam age m ap w ith an application to H u o n g R iver basin, in Central V ietnam
N guyen Tien G ia n g 1'*, Joric Chen2, Tran Anh P h u o n g1
1 C ollege o f Science, VNƯ U niversity o fT w e n te , The N etherlands
Rcccivcd 5 April 2009; rcccivcd in rcviscd form 18 April 2009
A bstract. In rcccnt ycars, uiidCT impacts o f human activities and cliniatc changc, flood has bcen increased in both thc ữcqucncy and magnitude, causũig lots o f damagcs to people. This study presents a m cthod to cvaluatc thc dữcct spatial damages o f ílooding bascd on inundation deptii and iand use data. D aniagc lunctions for diíTerent types o f land usc arc sclcctcd and applicd. Matlab strings and G1S arc com bincd to calculalc damagc in monitery tcrm. The mcthod is applicd for cstimating loss causcd by th e ũood event November 1999 for Huong rivcr basin. The rclative agreement bctwccn total dainagc o f survcy data and cstiniatcd rcsults shown lliut th e nicthodology provided in tliis study is applicablc. The niclhodology can bc uscd to dctcrminc llic flood-induccd economic loss for cost-bcncíìt aiialysis in the ílood control projccts.
Keywords: Inundation map; Flood dam agc map; Huong river basin.
1. I n tr o d u c tio n
E conom ic a c tiv itie s in ílo o d -p ro n e a re a s are b ein g d en tisiíied a ro u n d th e Nvorld. A t th e sam e tim e w e face c h a n g in g w e a th e r co n d itio n s a n d a rising sea level as a resu lt o f c lim a tic c h an g e . ỉ f no m easures a re c a rrie d o u t, b o th p ro b a b ility an d im pact o f ílo o d s w ill in c re a se d rastically [18]. In order to se le c t e ffe c tiv e ly th e m itig a tio n m easures, b esid es so c ia l c o n c e rn s, th e d ecisio n m akers sh ou ld b e in ío rm e d w h ic h m c a su rc brings m o re e c o n o m ic b c n e fit to th e area . In this sense, eco n o m ic d a m a g e c a u s e d b y tlo o d m u st b e estim ated. T h is is s u e has b e e n the interest o f m any stu d ie s. T h e F L O O D s ite ’s
Corresponđing author. Tel.: 84-4-2173940.
E-mail: giangnt@vnu.edu. vn
rep o rt (200 6) p ro v id ed a g u id e lin e fo r socio- econom ic flo o d d am ag e ev alu atio n [3]. T he report b u n d le d d iíĩe re n t a p p ro ac h es to flood dam ag e estim atio n. D am ag es are divided into m acro-, m eso - a n d m icro -sc ales a n d then dam ag e is estim ated b y u sin g the diíTerent dam ag e functions b a se d o n the level o f the flo o d event. V an d er V een an d L og tm eijer (2 00 4) exten d ed the k n o w n co n cep t abo ut dam ag e íu n c tio n s w ith th e indirect econom ic effects on the rest o f th e re g io n a l and national econom y [16]. T hey in tro d u ccd the d eĩm itio n
“v u ln e ra b ility ”, a ủ in c tio n o f dependence, red u n d an cy an d su scep tib ility . G enovese (2006) c a ư ie d o u t a d am ag e a sse ssm e n t for the 2002 flo o d ev en t in Prague, C zech R q ỉu b lic [5]. She used dam age curv es an d m a x im u m dam age 10
N .T . Giang et a i / VNU Ịournal o f Science, Earth Sãences 25 (2009) 10-19 11
valu es p ro p o sed by K o k a n d V an d er S ande [9, 14, 15] to d eterm in e th e d am ag e p e r squ are m eter. T hese reso u rces are o f use to understand the ío u n d atio n s o f th e d am ag e cu rve an d the asso ciated m axim um d am ag e value. V an der S an d e [15] created d etailed la n d c o v er m aps usin g satellite im ag es to im p lem en t for the d a m a g e ủ in c tio n s c o llected fro m [9] an d [11].
H uan g vvent into th e gap betvveen scientific kn o \v ied g e availab le an d its im p lem en tatio n o f d ecisio n su p p o rt sy stem , w h en riv e r b asin m a n ag em e n t ap p licatio n s w e re u sed [7]. Her thesis d ealt w ith the d ifficu lties re la te d to the selec tio n and p e río rm a n c e evalu atio n o f h y d rau lic m od els for F lo o đ R isk A ssessm en t (F R A ). A m o n g o th e r F R A 's , th e F R A using d ep th-dam age curves p ro p o s e d b y Kok, LKSR a n d V an d er S an de [4, 9, 14] w ere listed. The U.S. A n n y C o rp s o f E n g in eers d ev elop ed H EC - FD A m odel to fo rm u late a n d ev alu ate flood dam age red u ctio n p la n s u sin g R isk -b ased A n aly sis E x p ected A nnual D am ag e (EA D ).
EA D red u ctio n w as co in p u ted as th e d ifference betvveen EA D \vith a n d w ith o u t altem ativ e projects, a quan tity u s e d to a id in flo o d dam age reduction p ro jec t selec tio n [8].
In ta k in g th e ab ility o f G eo graphical In ío rm atio n S y stem (G ỈS ) in sp atial analysis, this stu dy attem pts to su g g e st a m eth o d to co m bine th e in u n d a tio n d ep th a n d la n d use daía to estim ate flo o d d am ag e m a p b y ad o p tin g the dam age fun ctio ns a n d p o te n tia l d am ag e values for differen t types o f land uses. W ith this m ethod, the d irect d a n ia g e p resen te d in m o nitery can b e determ ined. H ow ever, due to lack o f relav an t d ata, other ía c to rs like flood d u ration an d flo w v elo city o f ílo o d are not acco u n ted for, alth o u g h th e se factors m ay have im portant roles in ílo o đ d am ag e analysis. An ap p licatio n o f th e m e th o d to sim u latio n o f the N o v em b er 1999 ílo o d a t H u o n g riv e r b a sin w ith inu ndation d epth m o d e lled b y H E C -R A S and H EC-G eoRA S so íh v are is p resented in tliis paper.
T h e p a p e r is o rg a n iz e d as follows. In S ectio n 2, th e m e th o d o lo g y to co n stru ct dam age m ap is e x p la in e d In S e c tio n 3, the application o f th e m e th o d to H u o n g riv e r b asin are p resen te d . S e c tio n 4 is d ev o ted to the d ic u ssio n s o n th e se resu lts. F inally, in Section 5 co n clu sio n s o n th e m e th o d o lo g y and the o b ta in e d resu lts a re p resen te d .
2. M e th o d o lo g y
T h e c o n s tru c tio n o f flo o d dam age m ap re q u ire s h u g e a m o u n t o f d a ta related to flood a n d land. H o w e v e r> th e a v a ila b le data norm ally o n ly co n sists o f in u n d a tio n d ep th an d land use m ap. T h e re fo re d a m a g e íu n c tio n s should be a p p lie d b a s e d o n th e s e tw o ch aracteristics. The p ro c e d u re to d e v e lo p d a m a g e m ap is presented in F ig u re 1. T h e re a re fo u r types o f land use c o n sid e re d in th is stu d y . F o r each type, a d a n ia g e fu n ctio n in c lu d in g th e depth-dam age c u rv es an d th e m a x im u m d am ag e values is s e le c te d o r c o n s tru c te d . B y u n itin g flood depth w ith la n d use m a p s, th e la n d u se and inundation d q 3 th a t e v ery g r id cell c an b e determ ined. The d a m a g e íu n c tio n s , th e n , w ill b e ap plied to these c ells to e s tim a te e c o n o m ic lo sses. T otal dam age is a lso c o rrp u te d a s th e su m n ia tio n o f all grid cells in th e s tu d ie d area.
2 .1. D a m a g e j\'unction
T h e fu n ctio n c o n s is ts o f a dam age curve an d th e m a x im u m d a m a g e v alu e. T h e d q )th - d a m a g e c u rv e re p re s e n ts th e v u ln e ra b ility o f the c o n c e m in g o b je c t o r a s s e t vvhen it is Aooded.
T h e m a x im u m d a m a g e v a lu e is id en tiíĩed by th e lo ss v a lu e in c a s e th e o b je c t o r asset is fully ílo o d ed . M an y re la tiv e d ep th -d a m ag e functioiis e x ist in th e lite ra tu re . H o w ev er, because the d q )th -d a m a g e c u rv e an d th e m a x im u m dam age v a lu e d e p e n d s tro n g ly o n the characteristics o f
12 N .T . Giang ct ttỉ. / V N U Ịournal ọ f Sãetice, Earth Sáenccs 25 (2009) 10-19
the object o r asset, a dam age function sh ou ld n o t b e sim ply tak en o v er fro m a prelim inary research. T herefore, in th is study, dam age
fu nctio ns th a t have b c e n d o ne in the V ietnam area o r c o n stru cted in Ih e areas w ith sim ilar conditions to H u o n g riv er b a sin are selected.
Fig. 1. Flowchart for construcúng damagc map.
H o u seh o ld s D a m a g e: K o k [10] d evelop ed the S tandard M etho d to cstim ate flo o d dam age fo r v arious la n d uses in clu d in g houses. H e m ade d iíĩe re n t tu n ctio n s fo r differen t h o u sin g types fro m low -rise b u ilđ in g s (tw o m ain floors), m iddle-rise build in gs (fo u r m ain ílo ors) and high -rise b uild ings (six m a in ílo o rs). A v isit to the city o f H ue an d its siUTOiindings m ad e clear th at the urban area now aday s is íĩlle d w ith m ostly th ree o r four level b u ild in g s, w h ilc in the ru ral resid ential area houses w ith one o r tw o m ain íloors are m ore co m m on . S o for the urban residential area th e m id d le-rise b u ild in g curve w ill b e used an d for th e ru ra l resid en tial area the low -rise b u ild in g cu rv e w ill b e ad o p ted as the d epth-dam age curve.
T he u s e d curv es are sh o w n in F igure 3. In these cu rv es, the econ om ic dam ag e is the dam age o f th e b u ild in g p lu s the loss o f assets.
In th e im p lem en tatio n o f the S tan d ard M ethod o f th e d a m a g e c u rv e fo r houses, th e factor b etw een th e co n c e m in g m axim um dam age v alu es is 0.41.
E xtra íacto rs: T h e u s e d curves also proviđe the dam age fa c to r for th e su m o f the direct dam age, d irect dain ag e d u c to p rod uction loss a n d in d irect d am ag e. D u e to the substitution effects in th e su rro u n d in g s o f the ílo o d ed area, the dam ag e fu n ctio n for th e h ou seholds m ust be gi ven a ía c to r o f 0.75. M oreover, the area for urban resid en ts a n d ru ra l residents is not co m pletely íĩlle d w ith hoiises. A surface íactor is used to co rrect th is in form atio n. F actor o f 0.4
N .T . Giang et a i / V N U Ịourtial o f Sàe?ice, Earth Sàences 25 (2009) 10-29 13
is u s e d for the u rb an area w h ile ía c to r o f 0.2 is u s e d fo r the ru ral area. T h ese tw o íacto rs are im p lem en ted b y u sin g th e fo llo w in g íunction, in w h ich EED is th e eco no m ic dam age, d is dam age co rresp o n d in g to the in u n d a tio n depth in m eters, a is th e d am ag e fa c to r for the househo lds, Ỵ is the red u ctio n fa c to r an d p is th e su ría c e factor.
Ị ) h o u s e h o l ^ ^C cìuction/óor ^ f t s u r f a c e f đ o r ( 1 )
M axim um d am ag e value: S ince the dam age cu rve o f the Stanciard M eth o d o f K ok [10] is used, th e m a x im u m dam ag e v alue can be d evelo ped w ith th e fo u n datio ns fro m this m ethod. T he m a x im u m darnage values fo r a h o u seho ld are d efin e d as th e re b u ild in g cost for houses an d the re p la c e m e n t v a lu e fo r the assets inside th e house. A c co rd in g to T a b le 1, a v alue
2
o f 29 U S D /m is u s e d as n ia x im u m dam age value for houses in th e u rb an area w h ile a v alue
2
o f 22 U S D /m is u sed fo r ru ral area.
I n ỷ a s tm c tu r e dam age: In this study, in írasừ u c tu re is co n ce n trate d o n d am ag es cause for roads. D e B ru ijn [2] c o n stru cted a dam age cu rve fo r the roads a n d railw ay s in ử ie M ek ong D elta. T he curve is sirrũlar fo r d iffere n t types o f roads b u t the m a x im u m d am ag e v alu es are d ifferent for the highvvays, p ro v in c ia l roads and railroads. The d am ag e to ru ra l road s is n eg lected since n o m a x im u m d am ag e v alu e for this ty p e o f ro ad is know n.
E x ư a íactor: T h e reso u rce d ata o f the in írastru ctu re exists in U S D /m Hovvever, since the dam ag e íu n c tio n needs its in p u t in square m eter, a conversion facto r is u sed to co rrect the con version from a U S D /m u n it in to a Ư S D /rrf unit. T his factor is calcu la ted u sin g the averag e length o f a ro ad th ro u g h a cell a n d the area o f this cell. T h e ío rm u la fo r this ía c to r is g iv e n in Figure 2 . T h is ía c to r is th en im p lem en ted in the follow ing d am ag e form ula:
^ c o n v e r s io n ^"m lhistructurc ^
in w h ich E E D is th e econ om ic dam age in U S D /m ; infrastncture is th e dam age íacto r for the sp eciíĩc in frastru ctu re, d is the econom ic dam age co rresp o n d in g to the inu ndation depth in m eter, w h ich d e íĩn e d fro m dam ag e ũin ctio n an d Ciconvcrsion is th e co n v ersio n facto r from U S D /m to U S D /m 2.
Fig. 2. Calculation o f surfacc factor and inlrastructurc cclls (The lcngtli o f one cell is 90
meter).
M ax im u m d a m a g e value: T h e m axim um dain age v alue o f th e p ro v in cial road is 80 U S D /m w h ile th is v alu e for th e national ro ad is 4 0 0 U S D /m an d fo r the railw ay th e m axim um dam age v alue is 1000 U S D /m
Fig. 3. Used depth - damage curves.
Cmjg? ciívn
f i u i ể h o u M to td i, S lanổard $D0$}
ttoslNCtdre.Pa (ỉlfc) Rka l*nđ, Dt &vjp PX5)
Otop crof J. 0* Ei>ijn (2C05) fữrtx v n Htar.ỊŨ O X)___
ỉ 4 5 6 7
Inundittcp J*ptN jm)
14 N.T. Giang et ai. / V N U Ịoum al ọ f Science, Earth Sciences 25 (2009) Ĩ0 -Ì9
A g r ic u ltu r e dam age: D ue to th e sim ilarity in co n d itio n s, th e D c B ru ijn 's d am ag c cu rvc for ric e in th e area o f M ek ong D elta is used for rice la n d as w ell as other crop in this assessm ent. A s r e g a rd to the m a x im u m d a m a g e v alu es for the crop, sh e uses a m axim um dam age v alue o f 4 4 0 U S D p e r hectare for the other crops than rice an d 2 0 0 U S D p e r h ectare for rice.
D a m a g e f o r fo r e s t: T h e th e sis o f H u ang [7]
p ro v id e s th e d a m a g e c u rv e a s sh o w n in Figure 3 . R ep o rt o f C R U E IP [1] p ro v id ed the re p la c e m e n t v alu e for fo re st area in th e T hua T h ie n H u e P ro v in ce vvhich is equal to 0.84 U S D /m 2. T h is v alu e is a d o p te d as the m ax im u m d a m a g e v alu e.
Tablc 1. Used maximum damagc values pcr land usc typc
Uscd maximuni daniagc value Resourcc charactơislic ExƯa ĩactors
Urban area 29 USD/m2 Construction cosls & Surfacc factor: 0.4
Asset value Rcduction ĩactor 0.75
Rural area 22 USD/m2 Constraction costs & Surĩacc factor: 0.2
Asset value Reduction factor 0.75
Provincial road 80 USD/m Construction cosls Convcrsion ĩactor: 1/75
National road 400 USD/m Construction costs Conversion factor: 1/75
Railway 1000 USD/m Construction costs Convcrsion factor: 1/75
Rice 0.044 USD/I112 M aximum dam agc valuc
Other crops 0.02 USD/ m2 M aximum dam age valuc
Forest 0.84 USD/m2 Rcplaccm cnt valuc
3. A p p lic a tio n o f th e m e th o d to c o n s tr u c tio n o f d a m a g e m a p fo r H u o n g r iv e r b a s in
3.1. S tu d y area
T h e H u o n g R iver B asin is lo c ated in T hua T h ien H ue, a Central p ro v in ce o f V ietnam , b o rd e re d o n the east b y the E ast Sea, on the
\vest b y th e L aos. T h e H u o n g R iv er originates in th e m o u n tain o u s area aro u n d th e b o rd e r w ith L ao s an d flow s in the N o rth -E a ste m direction to the co ast. T h e H u o n g R iver B asin an d its a d ja c e n t area em brace a n area o f 3 7 6 0 km 2, o f w h ich 2 9 6 0 k m : b elo n g s to th e m ain H u o n g R iv e r B asin , an d the ren ia in in g o f 800 k m 2 b elo n g s to con tigu ou s basins. T h e H u o n g R iver flo w s in to a co n caten atio n o f lag oo ns n e a r Hue, fro m w h ich it leads to the E ast Sea. M uch o f th is p r o v in c e 's in frastructu rc an d in du stry lies in th e Coastal p la in an d m o st o f the popu lation s liv e w ith in 25 k m o f the coast. T his area has a sm all slo p e an d the H u o n g R iver an d its
trib u ta ry stre a m s m e an d er th ro u g h p opu latio n a n d a g ric u ltu re area.
T h e T h u a T h ie n H uc P ro v in ce has a tropical m o n s o o n c lim a te an d is a íĩe c te d b y annual ữ o p ic a l sto rm s. T hese ty p h o o n s usually d ev elo p in th e N o rth w e s t P acific an d follo w a p a th o v e r th e P h ilip p in es, c ro ss th e E ast Sea.
W h en lan d in g o n the V ietn am ese co ast they lo o se force, th e y release th e ir w ater o ver the Coastal z o n e [1 7 ]. B ecau se the H u o n g River B a sin is v e ry fla t in th e Coastal area an d the b a s in h a s n o suíT icient h y d rau lic stru ctures to h a n d le th is a m o u n t o f rain fa ll, it is under the h ig h ris k o f flood.
In N o v e m b e r 1999 a disasterous íloods s tru c k eig h t p ro v in c e s in C e n tra l V ietnam . Thua T h ie n H u e is o n e o f the p ro v in ce s that were a ffe c te d th e m o st severely. A p p ro x im ately 90%
lo \v lan d is u n d e r w ater. T h e flo o d lasted fo r one w eek, b ro k e fiv e new A oodgates an d created a n e\v riv e r m o u th n e a r the lagoon. N early a
N.T. Giang et a l ỉ V N U Ịourrtal o f Science, Earth Sciences 25 (2009) 10-19 15
m illio n ho m es w ere d am ag ed , o f vvhich m o re th a n 4 0 thousand w ere d estro y ed . T h e ílo o d in g c a u se d 265 m illio n U SD o f d arn ag e p lu s alm o st 5 0 0 m illio n U SD o f eco n o m ic lo sses [12].
3.2. D a ta a va ila b ỉe
Land use m ap: T h e land u se d a ta is o b ta in ed by th e V ie tn a m N atio n al U n iv ersity , H anoi fro m the g o v em m en t o f th e T h u a T h ie n H ue P rovince. A lth o u g h the d ata c o v ers th e w ho le province, only the d ata o f th e ílo o d e d a re a is disp lay ed sin ce the d ata o u ts id e the ílo o d e d area is not used. T he d ata has b e e n c o n v e rte d to ESRI files to view th e d a ta in ArcVievv so ftw are an d to co n v ert it to A S C II-file s w ith a 90 m eter grid. In F ig ure 4, th e 9 0 -m e te r g rid is disp lay ed w ith all the land uses th a t are u sed for the dam ag e m apping. T h e lo c a tio n o f th e left lovver co m er o f the lan d use íĩg u re is lo n g itu d e:
758.628.434 an d latitude: 1 .8 0 4.48 0.5 28 .
th e Bo R iv er in th e N o rth W est o f the figure n e a r the highvvay. T h e ru ral area is m ostly lo cated near the riv ersid e an d aro u n d the lagoons. It is su rro u n d ed b y rice land th at can b e fo u n d in the w h o le flat area o f the T hua T h ie n H ue P rovince. T h e ío rest land is located m o re uphill. The a rc a th a t is d cíìn ed ‘w ith other c ro p s ’ is th e land u s c d fo r ag ric u ltu re th at is not rice. B ecause in th e flo od ed area there are not m a n y o th er crops g ro w n , these are n o t d eíĩn ed in m ore detail. T h e ra ilro a d and h ig h w ay 1A ru n fro m the N o rth W est to th e S o u th East.
T hese ro ads are n o t dravvn in the íig u re from th e m om ent that they leav e the urban area, due to lack o f data.
Inundation m ap : T he inu n d atio n d q ) th m ap
\vith resp ect to th e m a x im u m w ater level at K im L o n g sta tio n d ev elop ed b y G ian g an d P h uo ng [6] w as used. In F ig ure 5, the sparial d ata o f the in u n d a tío n dq3tlì is draw n, w ith the locatio n o f th e le ft lo w er co m er, the g rid size a n d th e le n g th fro m top till b o tto m th a t is s im ila r to th e íig u re o f the lan d use. A s can b e seen in F ig ure 5, th e stud y area w as in un dated fro m the m in im u m o f 10 m m to m a x im u m o f 6078 mrrL
B ư M r are«
83 ss Mo^<r
^ yco -<fi aho KSrcMỉlcxi
Fig. 4. Land usc m ap o f thc study arca.
A s sh o w n in F ig ure 4, the u rb a n area is m ostly in H u e city th at can b e re c o g n iz e d b y the sq u are sh a p e o f th e c h a n n e ls a ro u n d o ld
H ue C itadel. A n o th er u rb a n a re a is lo c a te d a t Fig. 5. Inunđationm ap [6].
M Maximum: 6Q78 mm w. Mnimuin: 10 mm
16 N .T. Giang et al. / V N U Ịournal ọ f Sríence, Earth Sciences 25 (2009) 10-19
P opulation d en sity : T h e d ata o f the popularion d e n sity w e re o btain ed fro m the go v em m en t o f th e T h u a T h ien H ue P rovince. It consists o f th e to tal am o u n t o f resid ents per district, to g e th er w ith th e area o f each district.
This data, then, is u sed to calcu late the average density o f th e p o p u la tio n p e r district. T h e data are co nv ertcd to E S R I íĩles to v ie w th e d ata in A rcV iew so ftw a re an d to co n v ert it to A SC II- tĩles w ith a 9 0 -m e te r grid.
3.3. E stim a tio n o f ec o n o m ic d a m a g e
D am age m ap: B ased o n gi ven data, the dam age m ap o b ta in e d fo r th e H uong R iver B asin is sh o w n in F ig u re 6. T h e data is draw n in the cells o f 9 0 x 9 0 m 2. E con om ic dam ag e is presen ted in U S D /m 2. T h e sp atial dam age data in F igurc 6 sh o w s th a t th e in frastru ctu re causes the high est d a m a g e p e r sq u are m eter (17-24 U SD /m 2) w h ile ric e fields have the least dam age p er sq u a re m e te r w ith d an iag e b elo w 0.5 U S D /m . R em ark ab le p o in t is that H ue C ity has several lo c atio n s w ith a high p red icted dam agc, b u t in g e n e ra l it has on ly littlc dam age.
T his is p ro b a b ly cau sed b y the higher inu ndation d e p th d u e to the con v en tio n al irrigating s y s te m a ro u n đ th e o ld C itadel o f Hue.
F igure 7 p re s e n ts th e in u n d a ted area and dam age for diíTerent types o f lan d use. T h e ru ral area co v ers th e se c o n d largest inu nd ated area an d is s u b je c t to 53 p ercen t o f the total
dam age. T h e ra ilw a y s an d th e u rb a n land use are th e next la n d u se types th a t s u íĩe r large dam ages. T h e d a m a g e to th e fo re st area a n d to the area w ith o th e r cro ps is m arginal, the dam age to th e ric e fíelds w h ic h is not pro p o rtio n al to its in u n d ated s u ría c e are also negligible. In T a b le 2 , th e av e ra g e dam age p er lan d use ty p e is c a lc u la te d an d co m p ared vvith the used m a x im u m dam age. This sh o w s that the rice land, th e fo re s t a re a and a re a with other crops h av e a v e ra g e dam ages m o st close to its m a x im u m d a m a g e values. A fter th e se land use types, the ru ra l a n d u rb an areas h a v e th e highest dam ages in c o m p a riso n to its m axim um dam age values.
Fig. 6. Damage in USD/m2 using inundaied constructcd by HEC-RAS and HEC-GeoRAS.
N.T. Giang et al. / VNU Ịoum aỉ o f Science, Earth Sciences 25 (2009) 10-Ĩ9 17
* 1
Ỉ«MỊ* ti»A
■ÉỊÉÌVVíii'...
■ ■ 7 (UKM tgg»a s gg^l *- r«m:
1« w *
— • «•«*«?•
■m<5
Fig. 7. Damage and inundated area vvith rcspect to typcs o f land usc.
Tablc 2. Avcragc calculatcd dainagc and maximum daniagc valuc pcr laiid use
Avcragc damagc
M aximum daniagc valuc
Pcrcentage
Ricc 0,0403 0,044 92
Urbaii 5,53 29 19
Rural 2,67 22 12
Forcst 0,84 0,84 100
Highway 5,05 400 1,2
Provincial 1,05 80 13
roads
Railways 12,2 1000 1,2
OUier crops 0,0053 0,02 27
3.4. V alidation o f d a m a g e m ap
T h e total d am ag e acc o rd in g to th e dam age m ap w ith th e H E C -R A S in u n d a tio n m ap in this assessm en t is ab o u t m ore th an 2 0 0 m illio n U SD (see T a b le 3). U sin g the o b s e rv e d inu n d atio n m ap for a v alid atio n , th e to tal d am ag e is calcu lated to 28 5 m illion. C o m p a rin g this nu m b er w ith the d a m a g e rqD orted b y in tem atio n al reso u rces [13] w h ic h is equal to 265 m illio n U SD , the u sed d a n ia g e íunctions seem to b e q u ite accu rate. H ow ever, w hen in corpo rating the e ư o r c a u se d b y in undation
m ap th en th e d iffe re n c e b e tw e e n losses beco m es larg er (2 0 7 m illion s c o n p a r e d to 285 m illions).
Tablc 3. Total calculatcd daniagc and rcportcd dam ages in USD
HEC-RAS inundation map
Obscrved inundation map
Urban arca 36.679.000 66.589.000
Rural arca 110.220.000 154.590.000
Rice ficlds 4.096.000 6.383.800
Forcst area 2.633.400 2.612.700
Highway 17.207.000 20.627.000
Railroads 30.227.000 23.535.000
Provincial roads 5.746.100 10.859.000
Oứicr Crops 5.143 15.525
Total damagc: 207 nũllions 285 millions
4. D iscu ssion s
T h e re is n o d o u b t th at ílo o d dam age assessm en t co n tain s m an y u n certain ties Corning fro m b o th the m e th o d a n d its inputs. R egarding to input, this a sse ssm e n t assu m es th at the inu n d atio n d q ỉth d a ta is true. H ow ever, in the prev io us study, the sim p liíie d n c tw o rk w ithout
18 N .T . Giang et a i / V N U Ịourìíal ọ f Science, Earth Sciences 25 (2009) 10-19
co nsideration o f sm all ch an n els an d storages, the higher levcl o f roađ s a n d railw ay s in the study area reduces the accu racy o f inu ndation m ap. M oreover, other info rm atio n lik e flood intensity an d d u ratio n w h ich affect con sid erab ly on the dam age have n o t b een p ro vided. A s a result, dam ag e re la te d to th e se d ata is neglected.
R egarding to the land u se d ata, only a sm all am ount o f lan d use types is u sed a n d they can n o t represent all the ex istin g la n d use ty p es in the H uong R iver B asin. T his m eans th at n o t all thc p ossib le eco n o m ic d am ag e th a t th e H uong R iv er B asin suffers is taken into account. F or exam ple, industry, com m erce, tourism , íĩshery, recreational areas an d tem p les w o u ld b e subject to dam age. T h ese lan d u se types as w ell as the u sed land use types can cau se econoniic losses due to failure. S ince th is is not taken into acco un t in this assessm ent, this dam ag e m ap m ay create an inco m plete iin agc o f th e reality.
M oreover, the accu racy o f the v alid atio n is a íĩe c te d b y th e agc o f th c d ata o f th e lan d use w hich is o b tain ed recen tly an d flo o d event w hich dates b ack to 1999.
The u ncertainty o f d am ag e íu nction s is another source o f crror. A ll o f th c dam ag e curves u sed in this stu d y c o m in g fro m diíTercnt locations not H uong riv e r basin. A s a result, in som e cases, these fu nctio ns c an not re íle c t the relationship betvveen in u n d a ted depth an d econom ic losses. D etailed co n sid eratio n o f the dam age m ap is n o t v a lid a te d alth o u g h the to tal dam age o f the d am ag e m ap is validated. T h e spatial dam age results w ith in th e m ap actually a re in a black box.
5. C o n clu sỉo n s
In this p aper, a m eth o d to calcu late the dam age o f ílo o d is pro posed. B ased on inundation an d lan d use data, the m e th o d can
co nstructs th e dam age m ap b y u s in g dam age functions. S p atial an aly sis tech n iq u es o f GIS and codes o f M atlab are tw o m a in tools to q u an tiíy th e dam age. T h e relativ e agreem ent b etw een th e accu m u latio n s o f the d am ag e o f several la n đ u se types for ob serv ed in un d atio n d ep th w ith th e su rv ey đ an iag e p ro v e d that the con seq u en ces o f ílo o d in g on the H u o n g R iver basin can b e predicted b y the suggested m e th o d
F or the eco n o m ic dam age in case o f a flood like th e N o v e m b e r 1999 Aood, the a re a around rivers in th e H uong R iv e r B asin is a lso subject to th e la rg e st co nsequ ences. T his is m ostly caused b y th e settlem en t n e a r the riv ers and its larg e v u ln e ra b ility to ílooding. T h e rural area has th e la rg e st sh a re in th e total dam ag e and the second la rg e st area o f inundation, afte r the rice íĩelds. D eterm in in g th e d am ag e p e r square m eter, it ap p ca rs th a t the railro ad and the highw ays s u ffe r the m o st ex trem c dam age, follo w ed b y the iư b an settlem cnt. W hen ío rest area o r rice fields are stru ck b y inundation, its m a x im u m d a m a g c is reach ed vvith o nly a sm all vvater depth.
A ck n o v v leg d em en ts
T his p a p c r is p atially resu lted from the prọ ịect fu n d ed b y H ano i U niversity o f S cience (T N -07-50). T h e au th o rs w ould like to thank to thaĩ arrangem ent.
R e íe re n c e s
[1] CRƯEIP, Supplementary Appendix ío the Report and Recommendatìon o f the President to the Board o f Directors [Resettlement plan]. Central Region Urban Environmental Improvement Project. Provincial People’s Committee of Thua Thien Hue, 2003.
[2] K.M. De Bruijn, Resilience and fìo o d risk management, PhD thesis, TuIX‘lí\, DUP Science, Delíì University Press, 2005.
N.T. G ian% ct nỉ. / VNU Ịoum al o f Science, Earth Sàences 25 (2009) 10-19 19
[3] FLOODsite report, Integrated Risk Analysis and
Management methodology,
http:/ Avvvw OckxLsile.nct.. 2006.
[4] IKSR, Rhine Atlas 2001, Damage. Internationale Kommission zuni Schutz des Rheins, Druckbetricbe Letlner KG, 2001.
[5] E. Genovese, A meíhodologicaỉ approach to ỉand use-based flo o d damage assessment in urban areas: ỉ*rague cơse study'\ European Commission Dừectorate-General Joint Research Centre, 2006.
[6] T.G. Giang, T.A. Phuong, Application o f HEC- RAS and HEC-GeoRAS fo r /orecasting inundation map in Huong river basin, Final report o f the scientiíìc project TN-07-50.
Vieinam National ưniversity, lianoi, 2007.
[7] Y. Huang, Appropriatc modelling fo r integrated ỳlood risk assessmení, PhD thesis, University o f
Twente, 2005.
[8] Hydrologic Enginccring Cenưe Davis CA, HEC-FDA Sensitivity and Uncertainty Analysis, United States, 2001.
[9] M. Kok, Damage /unctions fo r the Meuse River Jloodplain, Communication paper to the Joint
Research Centre, Ispra, Italy, 2001, 10 pp.
[10] M. Kok, Standaardmethode 2005 Schade en Slachtơffers als gevolg van overstromingen (Standard Method), Rijkswaterstaat DWW, 2005.
[11] E. Penning-Rowsell, Stage-Damơge Functions fo r Natural Haiards Unit, F;lood Hazard
Research Centrc, Middlesex University, England, 2001.
[12] Reliefweb-Red Cross, Vietnam: Floods Situation Report No. ly International Pedcration o f Red Cross And Red Crescent Societies, 1999.
[13] Reliefwcb - USAID, Vietnam Fĩoods Factsheet
#/, FY 2000, Source: United States Agency for International Development, 18 November 1999.
[14] C.J. Van dcr Sande, River flood damage assessment using Ikũìios imagery. Natural Hazards Project-Floods, Joint Research Centre o f Ihe Europcan Commission, 2001.
[15] C.J. Van der Sande, A segmentation and classifìcation approach o f IKONOS-2 imagery fo r land cover mapping to assisí fỉood risk and fỉood damage assessment. International Joumal o f Applied Earth Observation and Geoiníòrmation, 2003.
[16] A. Van der Vecn and c . Logtmeijer, Economic Hotspots: Vỉsualừing Vuỉnerability to Flooding.
Natural Hazarcis 36, (2005) 65.
[17] H. Vermue, Flood modeling and measure assessment fo r Huong River Basin, Bachelor thesis, University o f Twentc, 2006.
[18] J. Wagcmakcr, J. Leenders, J. Huizinga, Economic valuation o f fỉood damange fo r decision rnakers in the Netherland and the lower Mekong rìver basin. Paper III. 3, HKV consultanls, Lelystad, Ihc Ncứierland, 2007.