Ramesh Maddamsetty1 S.Surya Rao 2 K.Manjula Vani3 T.Shivaji Rao4
---------------------------------------------------------------------------------- --------
1Ramesh M, Asst Professor, GITAM, Andhra Pradesh, India, m_rameshgitam@yahoo.co.in
2Dr.S.Surya Rao, Professor, GITAM, Andhra Pradesh, India, s_suryaraogitam@yahoo.co.in
3Dr.K. Manjula Vani, Professor, J.N.T.U, Hyderabad, Andhra Pradesh, India, ronilekha@yahoo.com
4 Prof.T.Shivaji Rao, Gitam University, Andhra Pradesh, India, shivajirao32@hotmail.com
Why Dams Fail? https://sites.google.com/site/ayyagaribhujangarao/whydamsfailhttps://sites.google.com/site/shivajirao32/damburstforpolavaram
http://sites.google.com/site/profshivajirao/polavaramdam-0http://sites.google.com/site/profshivajirao/polavaramdam-6
http://www.usbr.gov/gp/ecao/horsetooth_carter/horsetooth_safety_dams/mod/chapter1.htm
http://www.usbr.gov/gp/ecao/horsetooth_carter/horsetooth_safety_dams/mod/chapter2.htm
http://www.usbr.gov/gp/ecao/horsetooth_carter/horsetooth_safety_dams/mod/chapter4.htm
http://www.usbr.gov/gp/ecao/horsetooth_carter/horsetooth_safety_dams/mod/chapter5.
http://www.usbr.gov/gp/ecao/horsetooth_carter/horsetooth_safety_dams/mod/chapter6.htm
http://www.usbr.gov/gp/ecao/horsetooth_carter/horsetooth_safety_dams/mod/chapter7.htm
NOTE: All the above web sites and written comments upto this point are added to this paper by Prof.T.shivaji Rao and the other authors are not responsible for these additions which may provide more clarifications to the reader
Hundreds of dams failed all over the world
Many ignorant people and even engineers blindly believe that the dams do not fail at all.It is wrong.
Infact hundreds of dams failed in several countries including about 45 dams in India. Among the dams are the concrete dams, masonary dams, earth dams and rockfill dams. For details see the following websites.
http://www.usbr.gov/pmts/hydraulics_lab/pubs/DSO/DSO-98-004.pdf (See pages 56 to 60)
ABSTRACT
Drastic changes in geographical surface characteristics and meteorological characteristics lead to flash flood, whose magnitude if exceeds the capacity of spillways causes overtopping of embankment dams, resulting in the dam failure. When a dam fails, a large quantity of storage water is released to downstream, producing a flood wave which is capable of creating disastrous damage to the down stream people and property. Pre-determination of flood wave characteristics along the downstream river reach is very much essential in mitigating such disasters. The Prediction of characteristics of dam-breach flood wave formation and the downstream propagation for all the existing and the future major dams located in the earthquake prone areas and regions of heavy rainfall is very much essential. Accordingly the down-stream developments can be controlled, the possible extent of inundation of down stream zone can be predicted and the emergency action plan can be formulated to mitigate the disaster. The present study aims to predict the characteristics of the flood wave like peak flood stage, peak flood discharge and their times of occurrence at different locations downstream in the river due to dam-breach, for a hypothetical dam-breach pattern for a rock-fill dam on the Godavari River. The effect of variation of duration of breach of dam on the outflow hydrographs is also studied. The National Weather Service Dam Break Flood Forecasting (DAMBRK) model has been used for the study and the results are discussed in terms of outflow hydrographs.
1) Introduction
Dams have been playing a vital role in the development of any country by meeting the water demand for domestic use, irrigation, power generation, flood protection etc. There are about 45,000 large dams in the world. Lemperiere (1993) concluded that, today’s dams are ten times safer than fifty years ago, but population numbers downstream of dams have increased by a factor of twenty and are continuing to grow and such a growth of population is making the economical, technical criteria in determining the yardsticks of sustainable development. About 5% of the dams have been failing due to several factors like floods, landslides, earthquakes, deterioration of the foundation, poor quality of construction and act of war etc. When a dam fails or is deliberately demolished, large quantities of storage water are suddenly released, creating major flood wave capable of causing disastrous damage to downstream people and property.
2) HUNDREDS OF DAM FAILURES IN USA AND CHINA DUE TO MANY REASONS
According to the American Dam safety officers, out of 75,000 dams in USA, several hundreds of them are disasters waiting in the wings. Far too many dams are facing the risks of failure, threatening lakhs of human lives and billions of dollars worth of properties. Out of many old dams about 50-year old ones account for 85% of the dams by 2020 and most of these dams were built without adequate spillway capacities to release flood waters during torrential rains, causing extreme floods that overtop the dams ,resulting in their collapse
Even in China the Water resources Minister, Zing Ping recently stated that about 68 dams collapse every year. During the last 50 years about 3500 out of about 85,000 dams collapsed, placing dam collapse rate at 4%.In Guangdong province 50% of the dams amounting to 3685
are classified as Dangerous Reservoirs. Many cities are under threats of dam collapses and among them are 25.4% of cities with 179 dangerous reservoirs, and16.7% of county towns
with 285 reservoirs. In addition to 146 million people, about 9 million Hectares of cultivated fields also face serious threat. In fact in 1975,the collapse of Banquiao Dam caused death of 1,00,000 people due to drowning and 1,40,000 people due to the repercussions of the floods like epidemics and food shortage.
Predetermination of dam-breach flood wave propagation to the downstream river is very much essential. All the existing and the future major dams located in earth quake prone areas and region of heavy rainfall should be analysed for such a possible calamity, the flood due to the dam-breach should be routed downstream in the river, the possible peak river stages and river discharge-magnitudes and the times of occurrence with reference to the time of dam-breach should be computed. Accordingly the down stream developments can be controlled, the necessary engineering measures can be undertaken and the emergency action plan can be formulated to mitigate the disaster.
In the present study an attempt is made to predict the dam-breach outflow hydrograph for a major rock-fill dam of 5.4 billion cubic meter capacity which is proposed at Polavaram in Andhra Pradesh, India on the Godavari river, for a specified dam-breach pattern and the breach outflow hydrograph is routed through the downstream Godavari river reach of about 92 km from Dam site to the tail end of river which joins the bay of Bengal. The influence of variation of time of breach of embankment dams is also carried out.
The objectives of the study are:
1) To predict the breach pattern
2) To route the dam-breach flood superposed on the monsoon flood through the river system and determine the peak flood stage, peak flood discharge and their times of occurrence as a function of distance from the dam, stage and discharge hydrographs at different stations along the downstream river.
3) GOVERNING EQUATIONS
Some investigators used several existing dam-break models and concluded that the U.S National Weather Services (NWS) Dam Break flood forecasting model (DAMBRK) is reliable and well documented model. The governing equations of the model are the complete one-dimensional Saint-Venant equations of unsteady flow. The system of unsteady flow equations is solved by a non-linear weighted four-point implicit finite difference method. The 1-D Saint-Venant unsteady flow equations of conservation of mass and conservation of momentum are as follows:
in which, Q is the discharge; A is the active flow area; Ao is the inactive storage area; q is the lateral outflow; x is the distance along waterway; t is the time; g is the gravitational acceleration; h is the water depth; Sf is the friction slope; Se is the expansion-contraction slope.
4) DESCRIPTION OF THE CASE STUDY
The significant features of the rock fill dam considered in the study are as follows: The height of the rock fill dam above the deepest foundation level is 40 m and the upstream slope is 2.5H: 1V and the down stream slope is 2H: 1V. The length of dam at crest level is 2.3 km and the width of crest is 12.5 m. Length of reservoir is about 50 km. The gross storage capacity of the reservoir is 5,411 Mm3, in which live storage is 2,100 Mm3. The capacity of spillway is 1.02 Lakh cumecs. The maximum water level (MWL) is 53.0 m, Full Reservoir Level (FRL) is 45.72 m, spillway crest level is 30.0 m, Natural G .L is 13.5 m.
5) INPUT DATA
The rock-fill dam on the Godavari river is considered for the study. 1,000-year frequency discharge hydrograph is considered as inflow hydrograph to the reservoir. The spillway discharges corresponding to the reservoir head of water, the reservoir water level versus the reservoir volume capacity curve is considered for reservoir routing. Breach characteristics are determined by Froelich.DL (1995a) equations. The hypothetical cross-sectional details of down stream river reach are determined by Lacey’s formula for the available flood discharge data of the Godavari river basin [CWC(2006)] and used in routing the dam-breach flood, through the down stream river reach.( For tabulated input data, vide last page)
6) COMPUTATION OF THE BREACH DETAILS Breach initiates at a certain point on the top of the earthen or rock-fill dam due to overtopping. The breach widens and deepens and results in increasing flood flow through the breach. The predominant mechanism of breaching for earthen or rock –fill-dam is by erosion of embankment material by the flow of water over the dam. When the breach stems from overtopping, excessive shear stresses on the surface induced by water flow, initiates erosion process. Erosion will begin when local shear exceeds a critical value, after which earthen dam material is set in motion. The formation and duration of breach depending on the height of the dam, the material used for the dam construction, compaction of material, quantity and duration of flood flow. Overtopping breaches are usually either rectangular or trapezoidal in shape. The duration of breach is usually few minutes to few hours.
Coleman S.E, Andrews.D.P and Webby M.G (2002) carried out experimental studies on overtopping breaching of non-cohesive homogeneous embankments. The non-dimensional equations for prediction of breach pattern, from their studies are given a
Lb* = Lb/H = 16 (hb*)1.5 -------(3) H b* = (2.30 t* + 1) -1 -------(4)
Where Lb* = Lb / Hs, hb* = hb / Hs , Hb* = Hb / Hs, t* = gt2/Hsx106
Lb= length of breach crest, Hb = Height of breach crest above foundation,
g= acceleration due to gravity, t = time of breach.
Froelich.DL (1995a) studied 43 no. of breached dams of height ranging from 4.5 to 85.5 m and statistically derived predictors for breach pattern are given as
Bavg = 0.1803 ko (Vw.)0.32( Hb)0.19 --------(5) Tf = 0.00254 (Vw.)0.53( Hb)-0.9---------- (6)
Where Bavg = Average breach width, m; Tf = Time of failure, hrs; Ko = 1.0 for piping, 1.4 for overtopping failure modes, Vw = volume of reservoir, m3 ; hb = height of breach, m.
The author (2003) carried out least square analysis for 13 No. of historic medium dam failure cases and developed an empirical equation for average breach width (B), m and for the time of breach (T), hrs.
B= k (Hd)0.7 (vr)1.76 --------------(7) T= 26.28 (Hd)0.67 (vr) -0.264 ---------------------(8)
Where k = 2.78 x (10)-13, Hd= Height of dam, m Vr= volume of reservoir, m3
While using all the above breach predicting equations the calculated breach width from the Coleman (2002), and author (2003) breach equations are greater than crest length, but the calculated breach width from the Froehlich (1995a) breach equation is about 663m. Hence the breach width of 663 m is considered. In this study the calculated time of breach from the Coleman (2002), Froehlich (1995a) and author (2003) breach equations are about 7.38 hrs, 13.23 hrs and 0.84 hrs respectively. The time of breach of 0.84 hrs is considered in this study.
7) ROUTING OF THE BREACH FLOODa) NWS - Dam Break Flood Forecasting (DAMBRK) Model Description
The U. S National Weather Service (NWS) developed DAMBRK program (Fread, 1988) is reliable, well documented. The model has wide applicability, it can function with various levels of input data ranging from rough estimates to complete data specification, the required data is readily accessible and it is economically feasible to use with minimal computational effort on microcomputers.
DAMBRK model can be used to develop the outflow hydrograph from a dam breach and hypothetically route the flood through the downstream valley. The governing equations of the model are the complete one-dimensional Saint-Venant equations of unsteady flow, which are coupled with internal boundary equations representing the rapidly varied flow through structures such as dams and embankments, which may develop a time dependent breach. Also, appropriate external boundary equations at the upstream and downstream ends of the routing reach are utilized. The system of equations is solved by a nonlinear weighted four-point implicit finite difference method. The flow may be either sub-critical or supercritical.
The hydrograph to be routed may be specified as an input time series or it can be developed by model using specified breach parameters (size, shape, time of development). The possible presence of downstream dams which may be breached by the flood, bridge / embankment flow constrictions, tributary inflows, river sinuosity, levees located along the downstream river, and tidal effects are each properly considered during the downstream propagation of the flood. DAMBRK may also be used to route mud and debris flows using specified upstream hydrographs. High water profiles along the downstream valley, flood arrival times, and hydrographs at user-selected locations are the standard DAMBRK model output.
b) APPLICATION OF THE NWS-DAMBRK MODEL
The DAMBRK model is thoroughly studied. This model has been applied for several historical dam failure cases [Machhu dam-II, Gujarat, Teton dam,U.S.A] and satisfactory results have been obtained. The DAMBRK model has also been applied for a medium rock-fill dam (Rampada Sagar dam, Andhra Pradesh) to predict the maximum water levels along the down stream river for a hypothetical case of dam failure.
The NWS-DAMBRK flood-forecasting model is used in this study. The breach outflow hydrograph is predicted and routed through the downstream Godavari river reach over a length of 92 km. i.e. from Dam location to the tail end of river. The peak flood stage and the peak flood discharge profiles w.r.t distance from the dam are predicted, the times of occurrence of these quantities at different locations in the downstream river reach are also computed. The flood stage and the flood discharge hydrographs at different stations in the downstream river reach are also determined.
8) Results & DISCUSSIONS
It is assumed that when the Godavari River and its tributaries are simultaneously in flood state, and if the inflow flood to the reservoir exceeds the designed capacity of spillway, then the excess inflow flood overtops the dam. The overtopping of flood- water over the rock-fill dam initiates the breaching of the dam and large volumes of reservoir water releases into the river gorge and the breach flood wave propagates down the river in the downstream direction.
The dam-breach flood routing is carried out using NWS-DAMBRK model, for a hypothetical breach pattern for a rock-fill dam on the Godavari river basin. The duration of development of the breach from initiation to its final dimension due to overtopping flood is considered as 0.84 hrs and the consequent outflow through the breach and over the spillway is at its maximum value of 3,23,773 cumecs. The breach outflow flood is routed through the Godavari river system over a length of 92 km i.e. from dam location to the tail end of river. The resulting stage and discharge hydrographs at different locations on the downstream river reach are presented in Fig 3 to Fig 8. The predicted peak stage and peak discharge hydrographs w.r.t downstream distance from dam site to the tail end of river are presented in Fig 1 & Fig 2. The results of the sensitivity analysis of influence of duration of the breach of embankment dams on the peak flood discharge and on the peak flood depth of water are presented in Table 2 & Table 3, and Fig.9 & Fig.10 respectively.
The peak flood flow value of 3,23,773m3/sec, depth of water of 31.16 m occurred near the dam in 0.84 hrs from the commencement of breach. It takes further 6.34 hrs time for this peak flow to reach 30 km away from the dam and the peak flow at this section attenuated to about 2,56,320 m3/sec, and depth of water at this section is reduced to 20.29 m. It takes further 13.92 hrs time for this peak flow to reach 92 km away from the dam and the peak flow at this section attenuated to about 1,98,982 m3/sec, and depth of water at this section is reduced to 8.59 m. Similarly the peak flow, peak stage and their times of occurrence at different locations on the downstream river reach are tabulated in Table 1.
9) DAM BREAK ANALYSIS IN EIA REPORT BY STATE GOVERNMENT:
The Andhra Pradesh State Government , the project proponent submitted to the Central Government the Environmental Impact Assessment Report including the dam break analysis report prepared by the Experts of the National Institute of Hydrology, Roorkee in June 1999. The authors of the report emphasize that the objective of the report is to do hypothetical dam break flood analysis by preparing 1)input data of the study area compatible to the DAMBRK model, 2) the result in outflow hydrograph at various stations downstream of the dam and 3) the inundation map of the area. For the dam break flood the worst possible scenario is taken with the failure time as 30 minutes, breach length as 450 meters corresponding to the river width at bed level. The side slope of the breach is taken as 0.05 which corresponds to the slope of the Godavari river banks at the dam site. Dam-break occurs here by overtopping when the reservoir water level elevation is 53.32 meters that corresponds with the level at the top of the dam. The reservoir capacity at the top of the dam is calculated by extrapolation from the area elevation curve of the reservoir.
For the dam-break flood computation and routing NWS DAMBRK programme was used. Outflow hydrographs were taken for different sites at downstream distance of 6km, 12km, 20km and 30km respectively. As compared to the outflow peak flood of about one lakh cumecs (35 lakh cusecs) corresponding to passage of hydrograph over the spillway, this hypothetical dam break flood reaches a peak discharge of 1.56 lakh cumecs (55 lakh cusecs)at dam site and 1.42 lakh cumecs( 50 lakh cusecs) at 30km downstream of the dam at 0.5hours and 10.7 hour after the dam break respectively. Hence the peak flood for This dam failure will be one and half times the corresponding peak flood for no failure case, with the design flood as the inflow flood. A map with the boundaries of inundated area on both sides of the river banks are presented both for the spillway design flood for no-dam failure case and also for the hypothetical dam break flood of higher intensity. While the no-dam failure case design flood would pass through the river course and confined within the banks, the high intense flood due to This dam-break would inundate the lands for about 267sq.kms on the left bank side and 195 sq. kms on the right bank side.
Population likely to be inundated due to the collapse of This dam consequent to a maximum credible accident caused by extreme floods, earthquakes, human failures, construction defects, Dam collapses or sudden flood releases from dams in the upstream reaches of the river in other states etc.
POPULATION LIKELY TO BE DROWNED IN CASE POLAVARAM DAM FAILS
Town | Population | Mandals (E.G) | Population | Mandals (W.G) | Population |
Rajahmundry | 4,00,000 | Sitanagaram | 75,000 | Kovvuru | 70,000 |
Dowlaiswaram | 40,000 | Korukonda | 80,000 | Chagallu | 66,000 |
Mandapeta | 50,000 | Kadiyam | 85,000 | Nidadavolu | 70,000 |
Ramachandrapuram | 42,000 | Atreyapuram | 65,000 | Pentapadu | 72,000 |
Amalapuram | 55,000 | Mandapeta | 80,000 | Undrajavaram | 73,000 |
Kovvuru | 40,000 | Ramacharndrapuram | 70,000 | Tanuku | 70,000 |
Nidadavolu | 45,000 | Alamuru | 70,000 | Attili | 70,000 |
Tanuku | 70,000 | Ravulapalem | 80,000 | Ganapavaram | 70,000 |
Bhimavaram | 1,45,000 | Kottapeta | 80,000 | Akiveedu | 75,000 |
Palakollu | 80,000 | Kapileswarapuram | 70,000 | Undi | 65,000 |
Narsapur | 60,000 | Pamarru | 70,000 | Penumantra | 65,000 |
Yanam | 30,000 | Tallarevu | 80,000 | Penugonda | 70,000 |
I.Polavaram | 70,000 | Achanta | 65,000 | ||
Mummidivaram | 70,000 | Viravasaram | 65,000 | ||
Ainavilli | 65,000 | Bheemavaram | 80,000 | ||
Gannavaram | 75,000 | Mogalturu | 75,000 | ||
Ambajeepeta | 65,000 | Narsapur | 80,000 | ||
Mamidikuduru | 70,000 | Palakollu | 50,000 | ||
Razole | 70,000 | Elamanchili | 75,000 | ||
Amalapuram | 75,000 | Iragavaram | 70,000 | ||
Uppalaguptam | 62,000 | Palacoderu | 65,000 | ||
Rayavaram | 70,000 | Kalla | 70,000 | ||
Malikipuram | 75,000 | Poduru | 65,000 | ||
Sakinetepalli | 75,000 | Peravali | 70,000 | ||
Allavaram | 70,000 | ||||
Katrenikona | 75,000 | ||||
Total Population | 10,57,000 | Total Population | 18,92,000 | Total Population | 16,66,000 |
Government experts calculated that when the This dam collapses the devastating flash floods will enter Rajahmundry – Kovvuru region within 10 hours, with a flood depth of 20m. The floods enter in 14 hours in Tanuku, Ravulapalem, Mandapeta area within a flood depth of 17 m and Attili, Kothapeta, Draksharamam areas within 17 hours to a depth of 14 m and Naraspur, Amalapuram and Mummidivaram regions within 3 to 4 hours to a height of 10 m resulting in the Watery grave for about 45 lakhs of people in East and West Godavari districts. Thus a vast stretch of land between Kolleru lake and Kakinada containing hundreds of villages will be drowned causing deaths of about half a million people and few hundred thousands of cattle besides destroying the paddy fields covering more than a million acres of fertile delta lands. The submersion in the catchment area of the This reservoir is estimated at 2.3 lakhs of people who have to be rehabilitated in suitable areas that are provided with all buildings and other infrastructure facilities to protect their culture and quality of life.
Unfortunately the state government has not presented the comprehensive results of the dam break analysis report prepared by the Roorkee institute and hence they could not prepare the risk assessment and also the Disaster Management Plan as required under the provisions of the Environmental Protection act. The state cabinet and the people of Godavari delta were completely kept in darkness about the Environmental Impacts of the This project.
10) DAM SAFETY CRITERIA IN DIFFERENT COUNTRIES:In the case of the design of the dams the determination of extreme floods, inflow floods, safety check floods and spillway design floods are followed in different countries with different geographical, topographic and weather conditions in addition to ecological, sociological and economical criteria. The design floods for the spillways in the dams are based upon different criteria in countries like United States, United Kingdom and India as detailed in the following tables. The criteria followed for the case of this dam was taken by the A.P.State Government as 500 years return flood in 2005 and it was directed to be upgraded to 1000 year return flood. But in view of the emerging global warming impacts the Godavari catchment is expected to experience more intense storms of longer duration resulting in increase of floods by more than 20% and hence the design criteria for extreme floods in Godavari must be correspondingly raised.
In his latest article on design flood for dams F.Lemperiere,(International Journal on Hydropower & Dam, Issue2, 2005) stated that the discharge of extreme floods (such as the probable maximum flood) is in the range of 3 times the likely maximum discharge during the dams life. He stated that the failure of the dams by floods is caused by a small overtopping of the embankment dams and a huge overtopping of high concrete dams. The yearly probability of the design flood used for dams usually lies between 1/500 and 1/5000. He advocated from a realistic approach a “safety check flood” of very low probability (often chosen as the PMF), for which are accepted a reservoir level close to the crest of the dam and also some limited damages. He questions whether for answering the yearly probability of the maximum flood for ensuring safety of dam should be 1/1000 or 1/1,00,000 or quite nil. He states that the criteria for answering this question and the design methods are often the same as 50 years ago and have not been adopted to the present knowledge and conditions. He emphasizes that today there is much more data on extreme rains and floods which were considerably underestimated 30 years ago. According to him one of the most critical design criteria is that the volume and flow of an extreme flood (PMF) lie in the range of 2 to 5 or an average of 3 times the flow and volume of the maximum flood likely to happen during the life of the dam, i.e over 100 years. He says that the true return period of an estimated “1000 years flood” used as “design flood” may well be 200 years or 5000 years. He emphasizes that the design of mot existing dams are those under construction are based on a “design flood” which can be spilled (and possible partly stored) without damage. This International expert says that many small ungated embankment dams may withstand the peak maximum flood; but very large gated dams with a design flood of yearly probability of 1/1000 may fail for a 1/10000 flood with all gates open or for an yearly flood incase of all gates jamming. He further states that the evaluation methods are not the same for safety check flood which is close to the extreme floods and for the “operational flood” which is close to the 100 years flood. According to the reports on maximum reported flood collected from all over the world based upon catchment areas (ICOLD bulletin 125, page 75) the peak floods are presented :
Extreme flows reported worldwide| Catchment area, S(km)2 | 1 | 10 | 100 | 1000 | 10,000 | 1,00,000 |
| Flow (m3/sec) | 100 | 600 | 4,000 | 15,000 | 40,000 | 1,00,000 |
| Flow(m3/sec) per (km)2 | 100 | 60 | 40 | 15 | 4 | 1 |
| About 50 | About 2 | |||||
For S<300 km2, Q = 10,000(S/300)0.8
S>300 km2 , Q= 10,000(S/300)0.4
Comparison between dams of the same or similar region is reliable because the impact of the different soil and vegetation conditions is very similar including its shape and slopes.
http://www.hydrocoop.org/Shall_we_forget_the_traditional_design_flood.doc
USA STANDARDS – DESIGN FLOODS - HAZARDS
| S.No. | Dam | Hazard | PMF Value | Remarks |
| 1. | High | High | 1.0 PMF | |
| 2. | Intermediate | Moderate | 0.5 PMF | |
| 3. | Small | Low | 0.25 PMF | 100 years flood |
Note: Before 1900 design flow was based upon collection of data on high water marks on buildings and structures for calculating peak flood and spillways were designed by using a multiple of this known maximum flood as a factor of safety. But some dams failed because engineers used for spillway design the previous historical floods that are indicative of the maximum flood likely to be experienced by the dam during its design life.
RELATION BETWEEN “Q” FACTOR AND “RETURN PERIOD” OF FLOODS
| S.No. | Q (Cumecs, m3/s) | Factor | Return period |
| 1. | 36.3 | 1 (PMF) | 1,000,000 years |
| 2. | 18.2 | 0.5(PMF) | 10,000 years |
| 3. | 10.9 | 0.3 (PMF) | 1,000 years |
| 4. | 7.3 | 0.2 (PMF) | 150 years |
| 5. | 6.2 | 0.17 (PMF) | 100 years |
UK (1978) ICE GUIDELINES ON DESIGN FLOODS FOR DAMS IN TERMS OF PMF
| S.No. | Flood | Return period |
| 1. | 0.3 PMF | 1,000 year Return period flood |
| 2. | 0.5 PMF | 10,000 year Return period flood |
| 3. | 1.0 PMF | Category-A high dams with high hazard potential |
http://www.defra.gov.uk/Environment/water/rs/pdf/defra_rs_flood-etc-20.pdf
GUIDELINES FOR SELECTING DESIGN FLOODS, (CWC,INDIA)
S.No. | Structure | Recommended design flood |
1. | Spillways for major and medium projects with storages more than 60Mm3 | a) PMF determined by unit hydrograph and probable maximum precipitation (PMP) b) If (a) is not applicable or possible flood-frequency method with T = 1000years |
2. | Permanent barrage and minor dams with capacity less than 60Mm3 | a) SPF determined by unit hydrograph and standard project storm (SPS) which is usually the largest recorded storm in the region. b) Flood with a return period of 100 years (a) or (b) whichever gives higher value |
3. | Pickup weirs | Flood with a return period of 100 or 50 years depending on the importance of the project. |
4. | Aqueducts (a) Waterway (b) Foundations and free board | Flood with T = 50 years Flood with T = 100 years |
5. | Project with scanty or inadequate data | Empirical formulae |
Ref: CWC India “Estimation of Design Flood Peak”, Report No.1/73, New Delhi, 1973. | ||
11) Further StudIES
Dam-breach flood wave propagation models are very much dependent on the geometric and temporal dam-breach characteristics. The empirical equations presented in this paper for predicting the breach profile are to be refined by considering the data of several numbers of historical large earthen dam failure cases. The hypothetical cross-sectional data of down-stream river, which determined from the Lacey’s formulae are used in this study. The actual cross-sectional data of the Godavari river including flood plains on the down stream of dam may be considered for accurate simulation of dam breach flood characteristics. The sediment flow routing may also be considered to predict the riverbed profile
Acknowledgements
This work was carried out as a part of Research Project entitled “Dam Breach Flood Analysis” funded by Department of Science & Technology (SERC) New Delhi, India. The Funding of the study by the DST
(SERC), India, is gratefully acknowledged.
Table 1. Details of peakstage, peak flow and their times of occurrence at different locations on the downstream of dam (For Time of breach, Tf = 0.84 hrs).
INPUT DATA USED FOR THIS DAM ON GODAVARI RIVER
1 | 0 | 0 | 3 | 13 | 0 | 0 | 1 | |
1130.2 | 637 | 333.0 | 238.0 | 160.0 | 60.0 | 22.5 | 0 | |
55.0 | 45.72 | 40.0 | 36.0 | 32.0 | 24.0 | 18.0 | 13.5 | |
50.0 | 53.32 | 0.05 | 13.5 | 663.0 | 0.84 | 13.5 | 0.0 | |
53.32 | 53.32 | 0.0 | 35.72 | 0.0 | 9900.0 | 4000 | 1200.0 | |
2.0 | 120.0 | |||||||
170000 | 130000 | 90000 | 80000 | 70000 | 60000 | 50000 | 40000 | |
30000 | 20000 | 20000 | 20000 | 20000 | ||||
13 | 5 | 6 | 1 | 0 | 0 | 0 | 0 | |
1 | 2 | 3 | 4 | 5 | 7 | |||
0 | ||||||||
0 | ||||||||
13.5 | 20 | 34 | 40 | 60 | ||||
900.0 | 1350 | 1800 | 2350 | 2900 | ||||
0 | 0 | 0 | 0 | 0 | ||||
4 | ||||||||
13.0 | 16 | 21 | 32 | 40 | ||||
1000.0 | 1250 | 1800 | 3500 | 4500 | ||||
0 | 0 | 0 | 1000 | 4500 | ||||
6 | ||||||||
12.8 | 21 | 28 | 33 | 40 | ||||
1200. | 3100 | 5200 | 6000 | 8800 | ||||
0 | 0 | 0 | 2700 | 5000 | ||||
10 | ||||||||
12.2 | 22 | 28 | 34 | 40 | ||||
1350 | 3650 | 5900 | 7000 | 9800 | ||||
0.0 | 0.0 | 0 | 5000 | 7150 | ||||
12 | ||||||||
12.0 | 22 | 28 | 35 | 40 | ||||
1500. | 3800 | 6200 | 8300 | 10100 | ||||
0 | 0 | 0 | 5200 | 12900 | ||||
14.0 | ||||||||
11.8 | 21 | 27 | 34 | 40 | ||||
1600. | 4000 | 6500 | 8600 | 10500 | ||||
0.0 | 0.0 | 0 | 5200 | 15500 | ||||
20 | ||||||||
11.0 | 20 | 25 | 33 | 40 | ||||
1800. | 4400 | 8000 | 10800 | 13700 | ||||
0 | 0 | 1000 | 6000 | 15200 | ||||
30. | ||||||||
9.8 | 19. | 24.0 | 30 | 40 | ||||
2500. | 5000 | 9500 | 13000 | 16300 | ||||
0 | 0 | 3000 | 9000 | 18200 | ||||
52. | ||||||||
7.0 | 15. | 21.0 | 30 | 35 | ||||
5000 | 7500 | 10600 | 16000 | 20300 | ||||
0 | 2000 | 4500 | 11000 | 20200 | ||||
92. | ||||||||
2.0 | 11. | 15.0 | 30 | 35 | ||||
10000 | 12500 | 14600 | 20000 | 30300 | ||||
0 | 4000 | 8000 | 15000 | 20200 | ||||
0.035 | 0.035 | 0.05 | 0.05 | 0.05 | (Repeat 11 times more) | |||
0.50 | 0.50 | 0.50 | 0.50 | 0.50 | 0.50 | 0.50 | 0.50 | |
0.50 | 0.50 | 0.50 | 0.50 | |||||
-0.5 | -0.5 | -0.5 | -0.5 | -0.5 | -0.5 | -0.5 | -0.5 | |
-0.6 | -0.6 | -0.6 | -0.6 | |||||
0 | 0 | 0 | 0.00027 | 0.12 | 0.001 | 120. | ||
References
1) Choi, G.W., and Molinas, A. (1993) “Simultaneous Solution Algorithm for Channel Network Modeling”, Water Resource Research, Vol. 29, February, 321-328.
2) Coleman, S.E., Andrews, D.P., and Webby, M.G., (2002) “Overtopping Breaching of Noncohesive Homogeneous Embankments”, Jl. of Hydraulic Engg., ASCE,Sept, 829-838.
3) CWC (2006), Central Water Commission, Krishna and Godavari basin, Hyderabad.
4) Fread, D.L. (1998),DAMBRK: The NWS Dam-Break Flood forecasting Model, Office of Hydology, National Weather Service, Silver Spring, Maryland.
5) Froehlich, D.L.(1987), “Embankment Dam-breach Parameters”, Proceedings of 1987 conference on Hydraulic Engineering, ASCE, Aug.1987,570-575.
6) Froehlich, D.L.(1995a), “Embankment Dam-breach Parameters Revised”, Proc. 1995 ASCE Conf. on Water Resources Engineering, New York, 887-891.
7) Froehlich, D.L.(1995b), “ Peak outflow from Breached Embankment Dams”, Jl. of Water Resources Planning and management, ASCE, 121(1), 90-97.
8) ICOLD (1973), “Lessons from Dam Incidents”, abridged edition, USCOLD, Boston.
9) Kamalam., P.S.(2004) “Flood Routing in Tree Type of Channel Networks,” a M. Tech Thesis submitted to the Dept. of Civil Engineering, Andhra University, Visakhapatnam.
10) Lemperiere.F (1993), “Dams that have failed by flooding: an analysis of 70 failures”, Journal of Water Power and Dam Construction, September 1993, 19-25pp.
11) Nguyen, Q.K., and Kawano, H., (1995) “Simultaneous Solution for Flood Routing in Channel Networks,” Journal of Hydraulic Engineering, ASCE, Vol. 121, Oct pp. 744-750.
12) Ramesh.M and Praveen. T.V (2003), Dam-breach Flood Routing, Dr.of National Cont.on Hy & water.Res. HYDRO-2003, PP 45-48.
13) Ramesh.M, S.Surya Rao and K.Manjula Vani (2005), “Dam Breach Flood Analysis for High Rock-fill Dam”, Proc. of International Conference on Advances in Structural Dynamics and its Applications (ICASDA), December 2005, 345-365pp.
14) Ramesh.M, K.Manjulavani and Shivaji Rao.T (2006), “Dam Break Analysis as a Critical Parameter for Safety of Irrigation Projects – R.P.Sagar dam on Godavari river”, National Seminar on Disaster Management, Sri Venkateswara University, Tirupati, 27-28 February
15) Rao, K.L., (1995) “India’s Water Wealth,” Orient Longman Limited, New Delhi, India.
16) Satish Chandra, and Perumal, M (1985), Dam-break Analysis of Machhu Dam-II, Report of N.I.H, Roorkee, India.
17) Singh, V.P. and Scarlatos, P.D.(1998), Analysis of Gradual Earth Dams Failure, JI.Hydr.Div.ASCE, Vol.114, No.1, 21-41.
18) Surya Rao, S., S. Murthy Bhallamudi, S.K. Tewari, Ravi Bhushan Kumar, (2000) “Flood Routing in Tree Type Channel Networks, ISH Journal of Hydraulic Engineering, Vol. 6, No.1, pp 35-45
19) Tewari, S.K., (1996) “Flood Routing in Tree Type of Channel Networks,” a M. Tech Thesis Submitted to the Department of Civil Engineering, IIT Kanpur
20) Tony, L. Wahl (1998), “ Prediction of Embankment Dam breach Parameters”, Dam Safety Research Report, DSO-98-004, US Dept. of the Interior, Bureau of Reclamation.
21) Wurbs, R.E.(1987), Dam-break Flood Wave Models, Jl.Hy. Div.ASCE, Vol.113, 29-46.
Table 2. Details of peak flood discharge (cumecs) for different times of dam failure (Tf) and Times of occurrence (To, in hours) at different locations on the downstream of dam.
Table 3. Details of peak flood depth of water ( in meters) for different times of dam failure (Tf), and Times of occurrence (To, hours), at different locations on the downstream of dam.
ANNEXURE – I SALIENT FEATURES OF THE PROJECT
| Main works: | Design Flood | 0.102 M.cumecs | |
| Earth-rock fill dam | 2310 M long (7579ft) | Max. flood (1953) | 85,000 cumecs |
| Spillway in right flank | 906.50 M long (2974 ft.) | Annual rain fall | : 1022.95 mm |
| Power House in left flank | 9 Units of 80 MW each | Yield to be utilized Duty | 336.57 TM Cft 750 Ha/cum |
| Catchment area at head work site a) Gross b) Un-intercepted (between Polavaram and Pochampadu | 3,06,643 sq. km 2,15,957 sq. km) | Full Reservoir level Low water level (MDDL) Max. tail water level Min. tail water level | + 45.72 m (+150.00 ft + 41.15 m (+ 135.00 ft) + 30.48 m (+ 100.00 ft) + 13.64 m (+ 44.75 ft.) |
| Gross storage at FRL (145.72m) Storage at MDDL ( +41.15 m) Live Storage above MDDL (+41.15 m) | 5.111 TMCum (194.60 TMC) 3.381 TM Cum (119.40 TMC) 2.100 TM Cum (75.20 TMC) | Village Submersion: Andhra Pradesh Madhya Pradesh Orissa Total Villages | Nos 233 10 7 250 |
| Submersion (Lands) Andhra Pradesh Madhya Pradesh Orissa Total | Ha 44,513 1,504 1,026 47,043 | Length of dam Top of dam - level Average bed level Deep bed level | 2310 M +53.32 M 15.00 M +3.00 M |
| Top level of gates Crest level Size of gates No of gates | 45.72 m +25.72M 16M x 20M 44 | Deep foundn. level Spillway between abutments Left Canal Ayacut Right Canal Ayacut | (-) 6.10M 906.50 M 1.62 lakh ha. 1.29 lakh ha. |
Estimation of Probable Maximum Flood (PMF)
| Frequency (years) | Magnitude (cumecs) | Magnitude (Lakh cusecs) | |
| a. | 25 | 63,600 | 22 |
| b | 50 | 72,300 | 26 |
| c | 100 | 81,400 | 29 |
| d | 200 | 89,800 | 32 |
| e | 500 | 1,01,000 | 36 |
| f | 1000 | 1,09,400 | 39 |
ANNEXURE-II: INADEQUATE SPILLWAY CAUSES FAILURE OF DAMS (Cases in Gujarat):
Thus This project made of Earth and Rock-fill dam may be subjected to a maximum credible accident for various reasons. Moreover like so many dams which collapsed due to inadequate spillway capacities, This dam also has been designed about 30 years ago with highly inadequate spillway for discharging the peak floods. About 20 irrigation dams in India have collapsed. Even in Gujarat state several dams failed due to mistakes committed by the civil engineers in the design of the spillways as can be seen from the following table.
Design Floods, Actual Floods And Revised Spillways For Some Projects, Gujarat
| River Valley Projects in Gujarat | Total Catchment Area (sq.km) | Spillway Design Flood as per Project Report (cumecs) | Highest observed flood (Cumecs) | Revised Spillway (Cumes) |
| Dharoi | 5485.84 | 11213.00 | 14150.00 | 21662.00 |
| Dantiwada | 2862.00 | 6654.00 | 11950.00 | 18123.00 |
| Machhu-I | 735.00 | 3313.00 | 9340.00 | 5947.00 |
| Machhu-II | 1928.71 | 5663.00 | 16307.00 | 20925.00 |
| Damanganga | 1813.00 | 11100.00 | 12900.00 | 12854.00 |
| Source: Narmada, Water Resources & Water Supply Dept., Government of Gujarat http://www.sardarsarovardam.org/faqs/answers.htm#1 | ||||
Even in the case of Machchu dam failure the Morvi town was ill prepared to meet the 2 storey high wall of water that burst fourth from the dam 5 km upstream of the town and swept away the town on 11-8-1979 within a matter of 9 minutes. The waters receded after 4 hours and there were no emergency evacuation and disaster management schemes prepared as is the practice in USA for ensuring dam safety under the dam safety act that requires dam break analysis risk assessment and disaster management. Instead of the state Government officials at Rajkot the first news of the tragedy were known by the Americans who learnt about the dam collapse through the orbiting weather satellite much earlier than the Indian officials. The state officials admitted that the rainfall in the previous 24 hours was about 23 inches while the dam was designed to accommodate a maximum of 44 inches rainfall during the whole year.
ANNEXURE-III - INDIAN DAMS THAT COLLAPSED
| Dam | Type | Ht(m) | Years | Causes |
| Tigra (MP) | Masonry | 26 | 1914 - 1917 | Overtopping |
| Kundali (Mah) | Masonry | 45 | 1924–1925 | Structural |
| Pagara (MP) | Composite | 27 | 1927 - 1943 | Overtopping |
| L.Khajauri(UP) | Composite | 16 | 1949 – 1949 | Piping |
| Ahraura (UP) | Earth | 22.4 | 1954- 1955 | Piping |
| Kaddam (A.P.) | Composite | 22.5 | 1957 – 1958 | Overtopping |
| Kaila (Guj) | Earth | 26 | 1954 – 1959 | Piping |
| Panshet (Maha) | Earth | 53.8 | 1961 – 1961 | Piping |
| Kharagpur (Bih) | Earth | 24 | - 1961 | Overtopping |
| Kadakvasla (Maha) | Masonry | 40 | 1875 - 1961 | Overtopping |
| Kedarnala (MP) | Earth | 21.3 | 1964 – 1964 | Piping |
| Nanaksagar (UP) | Earth | 16.5 | 1962 -1967 | Piping |
| Chikhole(Kar) | Masonry | 36.8 | 1969 -1972 | Structural |
| Kodagnar (Tam) | Earth | 17.7 | 1977 – 1977 | Overtopping |
| Machchu (Guj) | Earth | 24.7 | 1972 – 1979 | Overtopping |
| Mitti (Guj) | Earth | 16 | 1982 – 1988 | Overtopping |
| Jamunia (MP) | 10 | 2002 | ||
| Lawa-Ka-bas (Raj) | 2003 |
ANNEXURE- IV - SELECTED CATASTROPHIC DAM FAILURE CASE STUDIES | |||||||||||||||
| S. No | Dam | Built- failed | Failure mode | Dam type | Height (m) | Length | Peak outflow (cu.m/sec) | Storage (M cu.m) | Vol above breach (M cu.m) | Depth above breach (m) | Breach height | Breach width (m) | Breach formation time (Hrs) | Failure time (Hrs) | Breach and empty time (Hrs) |
| (1) | (2) | (3) | (4) | (5) | (6) | (7) | (8) | (9) | (10) | (11) | (12) | (13) | (14) | (15) | (16) |
| 1. | Apeshapa (Colarado) | 1920-1923 | Piping | Earth | 34 | -- | 6850 | 22.5 | 22.2 | 28 | 31 | 93 | 0.75 | 2.5 | -- |
| 2. | Davis reservoir (California) | 1914-1914 | Piping | Earth | 12 | -- | 510 | 58 | 58 | 12 | 12 | 21 | -- | 7.0 | -- |
| 3. | Euclides (Brazil) | 1958-1977 | Overtopping | Earthfill | 53 | -- | 1020 | 13.6 | -- | 58 | 53 | 131 | 7.3 | -- | 7.3 |
| 4. | Hatch town (Utah) | 1908-1914 | Piping | Earthfill | 19 | 238 | 3080 | 14.8 | 14.8 | 17 | 18 | 151 | 1 | 3.0 | 1 |
| 5. | Johnstown (Pansilvania) | 1853-1889 | Overtopping | Earth & rockfill | 38 | 284 | 8500 | 18.9 | 18.9 | 24.6 | 24.4 | 95 | 0.75 | 3.5 | 3.5 |
| 6. | Kaddam (India) | 1957-1958 | Overtopping | Earthfill | 125 | -- | -- | 214 | -- | -- | 15 | 137 | -- | 1.0 | -- |
| 7. | Machchu dam | 1974-1979 | Seepage | Earthfill | 60 | 4180 | -- | 110 | -- | -- | 60 | 540 | -- | 2.0 | -- |
| 8. | Mammoth (USA) | 1916-1917 | Seepage | -- | 21.3 | -- | 2520 | 13.6 | -- | -- | 21 | -- | -- | 3.0 | -- |
| 9. | Nanak sagar (India) | 1962-1967 | -- | -- | 16.0 | -- | 9700 | 210 | -- | -- | 16 | 46 | -- | 12.0 | -- |
| 10. | Gros (Barzil) | 1960-1960 | Overtopping | Rockfill | 35.0 | -- | 9630 | 660 | 660 | 36 | 35 | 165 | 6.5 | -- | -- |
| 11. | Sallisoliveire (Brazil) | 1966-1977 | Overtopping | Earthfill | 35 | -- | 7200 | 26 | 71 | 36 | 35 | 168 | -- | 2.0 | -- |
| 12. | Teton dam (Idaho) | 1975-1976 | Piping | Earthfill | 93 | -- | 65120 | 356 | 310 | 77 | 87 | 151 | 1.25 | 4.0 | -- |
2 comments:
F.Lemperiere on Dam safety& Remedies
====================================
The increased safety now required for thousands of dams may be very costly if reached only by structural improvements. Various low cost non structural measures may reduce or avoid structural expenses and may be implemented in a short lime. They are even more attractive in developing countries.
The Committee on costs of ICOLD has prepared the bulletin E 102 on the subject: " Non structural risk reduction measures: benefits and costs for dams".
Based upon many comments and data from industrialized countries it has been written by M.Smart (U.S. Committee). This bulletin is freely available on the Net through: www.icold-cigb.org (Publications. Abstract of recent publications, Bulletins, E 02).
The present paper refers to many findings of this bulletin and analyses the efficiency of measures according to dam types and height and for various circonstances (first filling, floods, ageing...). It studies also specific data and advantages for developing countries.
This paper focuses on human risk which is actually the main basis of most criteria and decisions. For instance a number of countries require now to design or uprate the spillways for the probable maximum flood if the dam failure may cause fatalities. Applying this worldwide could modify dozen of thousands of dams and cost hundreds billons of $: this is unrealistic. But it seems possible to identify the true main risks, to reduce them on a number of dams by structural measures as far as they are cost effective and to insure a reasonable degree of safety by non structural measures applying to all dams at risk.
There are presently 45.000 large dams (39 000 being fill dams). Their failure would usually cause a wave flow of some thousands m3/s, and for 10.000 ofthem a flow over 10.000m3/s. Some past failures reached flows between 50 and 100.000 m3/s and possible failures could cause much higher flows. Moreover there are 100.000 small dams with storage between 0,1 and 3 millions m3 : their failure could cause waves in the range of 500 to 1.000 m3/s which may be dangerous locally in populated areas along small rivers.
As the world yearly rate of failure is in the range of 10-4 for the large dams and seems higher (possibly 5 x 10-4) for the small ones, some failures every year may have an impact on dozens
of km² and possibly over 100 km² : 50 or 100 may have an impact on few km² within a range of 10 or 20 km downstream of medium or small dams. The organization and cost of non structural measures may be different according to density of population (500/km² in most of Asia and 20/km² in most of America) and to failure hydro gram.
The efficiency of not structural measures may also vary according to dam type and failure circonstances because the time available for waming may be minutes or hours (or even days) and the percentage of fatalities in inundated areas may thus be less than 0,1 % or over 50 %. Night failures and cold water are evidently more dangerous.
Non structural measures include risk analysis, monitoring, training, maintenance, emergency planning and warning systems and modified operation
http://www.dailymotion.com/video/xfi1on_polavaram-project-ap-in-the-dock_news (Jairam Ramesh find faults with AP State Government for violating the Environmental conditions)
T.Sahithi Gupta
Post a Comment