The Rainstorm Which Caused the Morvi Dam Disaster in August 1979

7/27/2019 The Rainstorm Which Caused the Morvi Dam Disaster in August 1979

1/12

This article was downloaded by: [190.138.99.245]On: 30 September 2013, At: 17:39Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Hydrological Sciences BulletinPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/thsj19

The rainstorm which caused the Morvi

dam disaster in August 1979 / L'orage

qui a provoqu la catastrophe du

barrage Morvi aot 1979O. N. DHAR a , P. R. RAKHECHA a , B. N. MANDAL a & R. B.SANGAM

a

aIndian Institute of Tropical Meteorology, Poona-5, India

Published online: 21 Dec 2009.

To cite this article: O. N. DHAR , P. R. RAKHECHA , B. N. MANDAL & R. B. SANGAM (1981) Therainstorm which caused the Morvi dam disaster in August 1979 / L'orage qui a provoqu lacatastrophe du barrage Morvi aot 1979, Hydrological Sciences Bulletin, 26:1, 71-81

To link to this article: http://dx.doi.org/10.1080/02626668109490862

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (theContent) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoever as to

the accuracy, completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions, claims,proceedings, demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, in relation to orarising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden. Terms& Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions
http://www.tandfonline.com/page/terms-and-conditionshttp://www.tandfonline.com/page/terms-and-conditionshttp://www.tandfonline.com/page/terms-and-conditionshttp://www.tandfonline.com/page/terms-and-conditionshttp://dx.doi.org/10.1080/02626668109490862http://www.tandfonline.com/loi/thsj19


2/12

Hydrological Sciences Bulletin - des Sciences Hydrologiques, 26, 1, 3/1981

The rainstorm which caused the Morvi damdisaster in August 19790, N, DHAR, P. R, RAKHECHA, B. N, MANDAL &R. B, SANGAMIndian Institute of Tropical Meteorology,Poona-5, IndiaABSTRACT On 11 August 1979, the Machhu-2 earth dam,situated about 6 km upstream of the town of Morvi in theSaurashtra region of India, collapsed under the onrush ofan unprecedented volume of water. An 8-10 m high floodwave rolled down Machhu valley, entirely submergingMorvi and nearby villages. This flash flood caused thedeaths of thousands of people and totally destroyed urbanand rural property downstream of the dam. The heavyrainfall of August 1979 over and around the Machhu basinhas been analysed by both depth-area-duration and depth-duration methods. Important aspects of heavy rainfalldistribution such as: analysis of past severe rainstorms,maximum point rainfall of different return periods, andprobable maximum precipitation, were also studied. Thisstudy has shown that this event was not the most severerainstorm in this region. Possibly, the antecedentconditions of the Machhu basin played a significant rolein generating the flood volume which caused the earthflanks of the dam to give way.L'orage qui a provoqu la catastrophe du barrage Morviaot 1979RESUME Le 11 aot 1979 le barrage en terre Machhu-2,situ a environs 6 km l'amont de la ville de Morvi dansla rgion de Saurashtra aux Indes a t dtruit sous lapousse d'un volume d'eau sans prcdent. Une vague decrue de 8 10 m de haut a dval la valle de Machhusubmergeant entirement Morvi et les villages voisins.Cette crue brutale a caus la mort de milliers deriverains et a compltement dtruit toutes les propritsurbaines et rurales l'aval du barrage. La trs forteaverse de aot 1979 sur le bassin de M achhu et lapriphrie a t analyse la fois par la mthodehauteur-superficie-dure et par la mthode, hauteur deprcipitations-dure. Des caractres importants de ladistribution des trs fortes averses ont t galementtudis, tels que: l'analyse des orages svres dans lepass, la hauteur de prcipitation maximale ponctuellepour diffrentes priodes de retour et la hauteur deprcipitation maximale probable. Cette tude a montrque cet vnement n'tait pas l'averse la plus svreobserve dans la rgion. Il est possible que lesconditions sur le bassin, ayant prcd cet orage aient

71

Downloadedby[190.138.99.245]at17:3930September2013


3/12

72 O.N. Dharefa/.jou un rle significatif dans la formation de la cruequi a emport la digue en terre du barrage.

INTRODUCTIONThe Machhu basin located in the Saurashtra region (Fig. 1)experienced incessant rains during the first 12 days of August 1979.During this period, rainfall over Saurashtra region was more than7 times the normal for this period. Rainfall was, however,exceptionally heavy between 10 and 12 August, which caused hugeinflows of water into the two dams, viz., Machhu-1 and Machhu-2 (seeFigs. 1 and 2), located on the Machhu River at a distance of about30 km from each other. According to press reports, the Machhu-2earth dam was designed to pass a peak flood of 5.7 x 10 3 m3 s ,while on 11 August 1979 it was reported that a peak flood exceeding1.4 x 101* m s - was generated, which resulted in the over toppingof the earth flanks of the dam on the afternoon of 11 August 1979between 1430 and 1500 h 1ST (Fig. 3).

The worst affected area was Morvi town and neighbouring villagessome 6 km downstream of the dam, where a flood wave of 8-10 m inheight rolled down from the damaged M achhu-2 dam, submerging everything in its path. It is estimated that this flash flood caused theloss of thousands of lives, irreparable damage to livestock andproperty in the Morvi town which till then had been known as the"Paris of the East".This study examines the heavy rain spell of August 1979 overMachhu basin (up to Machhu-2 dam) and compares it with similar rain

Fig. 1 Map of Saurashtra region show ing isopleths of mean annual rainfall and probable ma xim umrainfall (PMP).



4/12

The Mo rvi dam disaster rainsto rm 737 1" 7 2"

69 " 70 71 7 2 Fig. 2 Isohyetal map of the 10 -12 August 1979 rainstorm over Machhu basin.

-./ . ^ **Fig. 3 Aerial photograph of the collapsed Machhu-2 dam . The earthen flank to the right of themasonry dam was washed away on 11 August 1979 and even 3 days after th e disaster the watercontinues to rush through the breached portion. (Photo by courtesy of the Information Department ofthe Government of Gujarat at Rajkot.)



5/12

74 O.N. Dharefa/.spells of the past 80 years in order to establish whether theAugust 1979 rain spell was unprecendented.

DATA USEDPast rainfall data for this region from 1891 to 1970 , available inthe archives of the Meteorological Office at Poona, were used inthis study. Daily rainfall data for the period 1-12 August 1979 forall the available rainfall stations in and around Machhu basin wereobtained through the courtesy of the Directorate of Agriculture,Government of Gujarat. Rainfall data for the meteorologicalobservatories were obtained from the State Meteorological Centre atAhmedabad.Before analysing the rainstorm of August 1979 some aspects ofheavy rainfall distribution in and around Machhu basin have beeninvestigated to establish the incidence of heavy rainfall in thisregion.

RAINFALL CHARACTERISTICS OF MACHHU BASINThe Machhu River, whose total length is about 140 km, rises in thehills of northern Saurashtra and flows into the Rann of Kutch. Itstotal catchment area up to Machhu-2 dam is of the order of 1900 km .Two important towns, Wankaner and Morvi (Fig. 1) are located on itsbanks besides numerous other big and small villages. Beyond Morvi,the Machhu flows through a flat treeless area and the river finallyloses itself in the Rann of Kutch. The normal monthly and annualrainfall of the entire basin up to Machhu-2 dam have been worked outby the isohyetal method (Fig. 1) on the basis of 1950 normals ofrainfall (India Meteorological Department, 1962) . Table 1 givesthe mean monthly and annual rainfall of the entire basin up toMachhu-2 dam; mean annual rainfall for the basin is of the order of569 mm, of which the southwest monsoon period (June-September)contributes about 533 mm. About 94% of the annual rainfall therefore occurs during the southwest monsoon season.Table 1 Mean mo nth ly and annual rainfall fo r the entire Machhu basin up to M achhu-2 dam (seeFigs. 1 and 2)

Months J F M A M J J A S O N D AnnualRainfall (mm) 1.3 2.1 2.3 1.8 9.2 84.9 249 .0 123.0 75.8 13.8 4.7 1.3 569.2% of annual 0.2 0.4 0.4 0.3 1.6 14.9 43.7 21.6 13.3 2.4 0.8 0.2 100.0

INCIDENCE OF HEAVY RAINFALL IN AND AROUND MACHHU BASINMaxim um observed 1,2 and 3-day rainfallFor long-period stations in and around Machhu basin, the maximumobserved 1, 2 and 3-day rainfalls have been extracted from the dailyrainfall records for the period 1891-1960 and are presented in



6/12

The Mo rvi dam disaster rainstorm 75Table 2. Table 2 shows that one station In this region, Palitana,experienced a fall greater than 500 mm in a single day. A t Morvi

Table 2 Ma xim um and extreme (PMP) rainfall values of stations in and around Machhu basin

Stations

BhavnagarBhujDhrangadhraGondalJasdanJunagadMandviMorviPalitanaPorbandarRahpurRajkotVeravalWadhwanWankaner

Meanannualrainfall(mm)62034050862461784440753461650035159452648756 1

Highest observed1891-1960 (mm)1-day307468441320278368311449508492184398302308308

2-day33050 151 251 2461679509533645625332475447445386

rainfall

3-day46350 769253954176855256967366035 2520529607452

50-yearrainfall(mm)25826933227 425337528328 7326415197309336234275

100-yearrainfall(mm)290307378308283422320324370474221348379264309

PMP I

593729838641568869679703834106344972379457464 7

and Wankaner stations the highest 1-day rainfalls of 449 mm and 308mm respectively were recorded on 23 July 1894. The 2-day highestrainfall for stations in the region varied from 33 0 to 680 mm andthe 3-day highest rainfall from 350 to 770 mm. It is alsoobserved that stations in this region have recorded their meanannual rainfall in a single day. A s an example Bhuj recorded 468 mmin 1 day in July 1959 which is 128 mm more than the mean annualrainfall for this station. D warka, a coastal station, recorded350 mm in 1 day in November 1951 which is almost equal to its meanannual rainfall (Rao, 1959).Point rainfall of 50 and 100-year return periods and probablemaximu m rainfall (PM P)Maximum 1-day point rainfall for 50 and 100-year return periods foreach of the long-period stations in and around Machhu basin havebeen determined using the G umbel (1954) procedure as adopted byChow (1964); these values of rainfall are also given in Table 2.The magnitudes of 50-year 1-day rainfall vary from 200 to 415 mm andthe 100-year from 220 to 475 mm.

Using H ershfield's (1961, 1965) statistical technique, which isone of the recognized WMO (1970, 1973) techniques for determinationof PM P, Dhar & Kamte (1971) prepared a generalized PM P chart for1-day duration for this region, using those stations which have60-70 years of rainfall data. The PMP pattern of this region alongwith the mean annual rainfall isopleths are shown in Fig. 1. Theestimates of PMP for some selected stations in and near Machhubasin, obtained from the generalized chart, are given in Table 2.One-day point PMP estimates in this region range from about 450 to1060 mm.


13


7/12

76 O.N. Dharef a l,ANALYSIS OF PAST SEVERE RAINSTORMSThe data for severe rainstorms for the period 1891-1970 over andnear M achhu basin were examined and it was observed that this regionexperienced the following three severe rainstorms during thisperiod: (a) 21-23 July 1894, (b) 25-27 July 1927, and (c) 12-14 July1950. It was observed that these three rainstorms had their centresoutside but close to M achhu basin. A ll three were analysed by thedepth-area-duration (DAD) method and maximum rain depths up to anarea of about 13 00 0 km 2 are given in Table 3. This tableindicates that the rainstorm of July 1927 was the most severe for 2and 3-day durations while the July 1894 rainstorm ranked second inorder of severity. The July 1927 rainstorm, with its centre nearTable 3 Depth-area -duration (D AD ) statistics of past severe rainstorms over and near Machhu basin

Area(km2)

Point25951811 11295*19002590518012950

July 18941-day(mm)450446436428417403393363303

2-day(mm)533527521519513507500477420

3-day(mm)543530526522516509502478423

July 19271-day(mm)457452437420392374365336281

2-day(mm)749718708698678665650604534

3-day(mm)841825816806794782770743682

July 19501-day(mm)399390376360338319304274229

2-day(mm)511510504498487474460420361

3-day(mm)538536532530522516507472409

*Ma chhu basin up to Machhu -2 dam has an area of the order of 1900 km .Dakor (2245'N, 7309'E) in north G ujarat, was analysed and it wasfound that this rainstorm gave areal rain depths which have not sofar been exceeded in this country (Dhar et al., 1980). This veryrainstorm moved westwards 2 days later and was centred close to theM achhu basin. The maximum rain depths at its new location aregiven in Table 3.

RAIN SPELL OF AUGUST 1979 OVER A ND NEAR MA CHHU BASINMeteorological situations responsibleStudy of meteorological situations during the first two weeks ofAugust 1979 has revealed that widespread rainfall in and aroundMachhu basin was mainly caused by the following synoptic situations:

(a) D uring the first week of Aug ust , two low pressure systemsmoved westnorthwest in quick succession along the monsoon trough,which was then lying south of its normal position. A lso, from 5 to8 A ugust, a cyclonic circulation extending to the middle tropospheredeveloped over G ujarat region. In association with these , themonsoon was fairly active over Machhu basin during the period 1-8A ugust causing widespread rainfall.

(b) There was another intense rainspell in and near M achhu basin



8/12

The Mo rvi dam disaster rains torm 77in the second week of A ugust 1979. This was caused by the movementof a low pressure area which developed in the northeast Bay ofBengal on 5 A ugust. It intensified into a severe cyclonic storm andcrossed the coast on the midnight of 7 A ugust. Continuing to movewestnorthwest it weakened into a low pressure area over Gujarat andadjoining southeast Rajasthan on 11 August, and subsequently mergedwith the seasonal monsoon trough by about 14 A ugust. The track ofthis disturbance up to depression stage is shown in the inset ofFig. 2. Under its influence, the monsoon was strong to vigorous inand around Machhu basin.

It is seen that during the rainspell, R ajkot, a station locatedclose to Machhu basin, recorded rainfall of 354 mm on 11 August1979 with return period about 10 0 year s. Maximum 1-day rainfalls atMorvi and Wankaner stations were found to have return periods of theorder of 25 year s.Average daily rain depths at Machhu basin during 1-12 August 1979As mentioned earlier, stations within Machhu basin experienced lightto heavy rainfall from 1 to 12 A ugust. During this period, rainfallwas particularly heavy from 10 to 12 A ugust. The average basinrainfall was calculated by the isohyetal method up to Machhu-2 dam,for each day of the rain spell; these data are given in Table 4.

Table 4 Average daily rain depths over the Machhu basin up to Machhu-2 dam during 1 -12 August1979

Date En tire basin rain Percentage of mean Percentage of mean(Aug ust, 1979) depths up to Machhu-2 Augu st basin rainfa ll annual basin rainfa lldam (mm)39573972332440222275237135

(%)32463259272033181861193110

(%)71071364744134224

Table 4 shows that the peak period of heavy rainfall over Machhubasin (up to Machhu-2 dam) was from lO to 12 A ugust. It is alsoseen that this basin received maximum rainfall of 237 mm on11 A ugust , which is nearly twice the mean rainfall for A ugust. Thetotal 3-day rainfall experienced by the basin from 10 to 12 Augustwas 447 mm, nearly 4 times the amount the basin normally receives inthe entire month of A ugust. Table 4 suggests that the antecedentconditions of Machhu basin, at the time of intense rainfall of 10-12 A ugust , were quite favourable for causing maximum runoff , as thebasin was almost saturated from the light to moderate rain of theprevious 10 days.



9/12

78 O.N. Dharefa/.Depth-area-duration (DAD ) analysis of 10-12 August 1979 rainfallover Machhu basinDepth-area-duration analysis of the 10-12 August 1979 heavy rainfallwas also carried out and DAD rain depths obtained for areas up to13 OOO km 2 are given in Table 5 and the 3-day isohyetal pattern ofthe rainstorm is shown in Fig. 2. This figures shows that the rainstorm had its heavy rain centre quite close to R ajkot which is about20 km west of Machhu basin (Fig. 2 ) . H ad this centre of heavy rainoccurred over the basin itself, perhaps the magnitude of the floodgenerated would have been greater and the disaster caused by thedam failure would have been even worse.

Table 5 Depth-area-duration (DAD ) statistics of the 10-1 2 August 1979 rainstormover Machhu basin

Area 1-day 2-day 3-day(km 2 ) 11 Aug. 1979 11-12 Aug . 1979 10-12 Aug. 1979(mm) (mm) (mm)Point 352 554 704259 350 552 6601295 336 535 6241900 330 525 6122590 326 518 60012950 266 425 504

The D A D analysis of the 10-12 A ugust rainstorm (see Table 5) hasrevealed that Machhu basin up to M achhu-2 dam could have experienced1-day, 2-day and 3-day rain depths of the order of 330, 525 and 612mm respectively had the rainstorm centred over the basin in acritical pattern.

COMPARISON OF DEPTH-DURATION (DD) VA LUES OF SEVE RE RAINSTORMS OVERMACH HU BASINThe average rain depths actually experienced by Machhu basin (up toMachhu-2 dam) during the severe rainstorms o f 1894 , 1927, 1950 and1979 were also worked out for durations of 1 to 3 days and thesedata are given in Table 6. This table shows that M achhu basinexperienced the highest rain depth for 3-day duration during theA ugust 1979 rainstorm. This is perhaps due to the fact that the

Table 6 Depth-duration (DD) rain depths for 3-day duration experienced by Machhu basin (up toMachhu-2 dam) during severe rainstorms

Dura tion 21-23 July 1894 25-2 7 July 1927 12-14 July 1950 10-1 2 August 1979(mm) (mm) (mm) (mm)1-day 280 229 22 9 2372-day 399 302 351 3723-day 400 387 377 447



10/12

The Morvi dam disaster rainstorm 79centre of the A ugust 1979 rainstorm was close to the basin.

TIME DISTRIBUTION OF AUGUST 1979 RAINSTORMThere are no recording raingauge stations located within the basinand as such the actual time distribution of the heavy rainfall of10-12 A ugust 1979 could not be assessed accurately. However,recording raingauges at Bhavnagar, Porbandar and Veraval stations,which lie in the vicinity of the rainstorm area, functionedsatisfactorily during the storm period. Hourly rainfall data ofthese stations were, therefore, used to determine the timedistribution of the A ugust 1979 rainfall. Figure 4 shows the timedistribution curves of these stations for 2-day rainfall, asrainfall of 2 days more or less represents 48-h rainfall (US WeatherBureau, 196 1, 1964) . By way of comparison, the time distribution ofAhmedabad station (Changraney & Jain, 1978), located east of Machhubasin, has also been shown in Fig. 4. This figure shows that thePorbander curve almost envelopes all other curves of this region andhence the time distribution curve of this station has been used towork out short duration rainfall of the basin. The maximum D Drain depths for the August 1979 rainfall (Table 6) for differentdurations are given in Table 7.

Table 7 Ma ximu m basin rain depths for short durations for the 10-1 2 August 1979rainstorm

HoursRainfall (mm)

1141

3167

620 8

1226 8

18312

24342

36364

48372

72447

o1 6 0


11/12

80 O.N. Dhar et a l .(a) there have bee n instan ces when sta tions in this re gio n

received their resp ective mean annual rainfall just in a single day ;(b) ma xim um 1-day ra inf all at these station s var ied f rom 18 0 to

510 mm and 2-day from 330 to 680 mm; and(c) PM P o f i-day durat ion worked out to be of the order of

4 5 0 - 1 0 6 0 m m .D uring the period 1 0 -12 A ugust 1979 , M achhu basin up to M achhu-2

dam received abou t 4 times the normal basin rainfall for A ug us t.D epth -dur atio n analysi s of pas t rainsto rms over this basi n has revealed that the July 1894 rainstorm gave maxi mum rain depth s fordurat ions of 1 and 2 days while the 10-12 A ugus t 1979 rainsto rm gavemaximum rain depths for 3-day duration. Depth-area-durationanaly sis of the past severe rains torms of this regi on has shown tha t,by and lar ge, the 1927 rainstorm was the mo st severe and therain depths measu red in this storm were much higher t han the rain storm of 1 0-12 A ugust 1979. A pparently , the large volume of watergenerat ed during the rains torm of A ugu st 1979 was mai nly due to thefavourab le anteced ent basin condit ions of mod era te to heavy rainf allduring the 10 days befo re the dam f ailur e.

A C K N O W L E D G E M E N T T h e a u t h o rs a r e g r a t e fu l t o D r B h . V . R a m a n a m u r t y ,D irecto r , Indian Institute of Tropical M eteorol ogy , for hisencou ragem ent and keen interest in the hydro mete orolo gical stud ies.Thanks are also due to M ess rs A . K. Kul karni and G . C. Gh ose of theH ydrom et R esear ch Proj ect for their help in the compilation andcomputa tion of data and to M r V . V . Sav argaon kar for typing thep a p e r .

REFERENCESC hangraney, T. G. & Jain, B. C. (1978) Study of time and areal

distribution of extreme rainfall for evaluating standard projectstorm values for small and medium catchments. Proc. Symp. on theHydrology of Rivers with Small and Medium Catchments, vol. 2.CBI&P Publ. no. 130 .

C how, V. T. (1964) Handbook of Applied Hydrology. McGraw Hill,New York.

D har, O. N. & Kamte, P. P. (1971) E stimation of probable maximumprecipitation over the coastal states of Gujarat and Orissa.Proc. Symp. on Water Resources. Indian Institute of Sc ience,Bangalore.

D har, 0. N., Rakhecha, P. R. & M anda i, B. N. (1980) R ainstorms whichcontributed greatest areal rain depths in India. Arch. Met.Geophys. Bioclim. ser. A , 29 (1-2).

Gumbel, E . J. (1954) A pplied M athematics S eries, No. 33 , NationalBureau of S tandards, Washington, D C .

Hershfield, D. M. (1961) Estimating the probable maximumprecipitation. J. Hydraul. Div. ASCE 87 (HY5).H ershfield, D. M. (1965) Method for estimating probable maximumrainfall. J. Am. Wat. Wks Ass. 57 (8).

India Meteorological D epartment (1962) Mem. India. MeteorologicalDepartment XXXI, Part III.



12/12

The Morvi dam disaster rainstorm 81Rao, K. N. (1959) Heavy rainfall of India. Proc. Symp. onMeteorological and Hydrological Aspects of Floods and Droughtsin India. India Meteorological Department Publication.US Weather Bureau (1961) Tech. Pap. no. 40.US Weather Bureau (1964) Tech. Pap. no. 49.WMO (1970) Guide to Hydrom eteorological Practices. WMO no. 168.Tech. Pap. 82.WMO (1973) Manual for Estimation of Probable Ma ximum Precipitation.WMO no. 332.Received 19 Ma y 1980; revised 8 July 1980.


Documents

The Rainstorm Which Caused the Morvi Dam Disaster in August 1979