International Journal of Engineering Research ISSN:2319-6890(online),2347-5013(print)

Volume No.5, Issue Special 1 pp : 207-209 8 & 9 Jan 2016

Dam Break Analysis - A Case Study

S. R. Kulkarni1, S. K. Ukarande2, Seema A. Jagtap3

1, 2
Civil Engineering Department, Yadavrao Tasgaonkar College of Engineering & Management,University of
Mumbai, Maharashtra, India
3
Civil Engineering Department, Thakur College of Engineering & Technology,Mumbai, Maharashtra, India
Email: 1sanjayramkulkarni@yahoo.com, 2ukarande@yahoo.com, 3 jagtap.seema8@gmail.com

Abstract: This paper is mainly focused on the effects of post 1.3. Objective:-
dam break. It can be estimated from the flood developed from “Pawana dam” which is gravity dam constructed on Pawana
the dam break. It will also throw light on high flood areas in River near Lonavala in 1972 is selected for case study of dam
the downstream side of the dam which will help assess the break analysis. Pawana Dam is located about 7 km from
type of emergency facilities. Lonavala and about 19 km in the interior of Kamshet. Pawana
Keywords: Dam-Break Analysis, Dam Breach, Dam CFD, Dam is has been used for generating electricity and supplying
Pawana Dam water for irrigation.
I. In tr oduct i on The dam break analysis willmake possible to estimate the
1.1. Dam Break Analysis:- flood and flood affected areas at downstream due to breach.
Dam break analysis is characterization and identifying of This enablesthecost estimation in case of rehabilitation.This
potential dam failures are post effects of resulting floods from study seesthe possibilities of precautionary measures which
dam breach. It is this characterization of the threat to public can be taken to completely avoid the dam break which avoid
safety that a dam poses decides the classification of the dam or minimize damage.
and also the standard of care, safety and maintenance to which This study will help the town-planner decide on the no
the dam is held. The requirement to prepare an emergency development zone, flood control lines within certain limit
action plan, requiring preparation of inundation maps which after the embankment along the river.
accurately predict dam breach flood depths and arrival times at
critical locations. The population associated along with the II. Methodology
critical section are located in close proximity downstream of a For understanding dam-breach analysis tools, it is important
dam, details of the breaching process and the calculated peak to understand the critical breach parameters. There are four
discharge may have little effect on the results. The breach such critical parameters:
parameters like breach width, depth, and rate of development 1) Breach parameter estimation (breach size/shape and
are more crucial to analyse especially when the locations of time of failure),
population centres are near to the dam. The associated cost and 2) Breach peak discharge and breach hydrograph
assumptions increases if the breach parameters cannot be estimation,
predicted with reasonable accuracy. A recent query of the dam 3) Breach flood routing, and
safety engineers within the Colorado Dam Safety Branch 4) Estimation of the hydraulic conditions at critical
determined that there is currently no consensus nor up-to-date locations.
guidance regarding the state-of-the-practice procedures for
performing dam breach analysis.This study can also estimate Empirical Methods:-
high flood level and, to fix the flood control lines on Empirical methods are used to predict time to failure and
downstream side of the dam, estimation of wave pressure after breach geometry, as well as to predict peak breach discharges.
dam break. The study also focus on to understand the hydraulic The empirical approach relies on statistical analysis of data
characteristics and breach shape, understand the flow of huge obtained from documented failures. The four most widely
debris from upstream as well as downstream which will be used and accepted empirically derived enveloping curves
carried away with flow after dam break, study the erosionof and/or equations for predicting breach parameters are:
dam after dam break, it give important information for MacDonald & Langridge – Monopolis (1984), U.S. Bureau of
embankment on the downstream sides of the river. It also helps Reclamation (USBR-1988), Von Thun and Gillette (1990),
to avoid erosion of soil along the embankment of downstream. and Froehlich (1995a, 1995b, 2008).[i]
1.2. Need for Dam-Break Analysis:- In this study Empirical equations have been used to estimate
the dam breach flood.
Two major consequences of a bam failure are:- A study on estimation of flood using Empirical methods was
1. Life loss: This loss occurs if the villages and the residing performed.
families are washed away by the flood resulting from dam- As mentioned in Guidelines for dam breach analysis the
break. MacDonald & Langridge-Monopolis (1984), Washington
2. Economic:Economic loss is calculated in terms of revenue State (2007) and Froehlich (2008) methods are the
which will be required to rebuild the washed away villages in recommended empirical tools for predicting dam breach
terms of infrastructure, and other allied facilities. parameters within the State of Colorado.

NCICE@2016 doi : 10.17950/ijer/v5i1/049 Page 207

 International Journal of Engineering Research ISSN:2319-6890(online),2347-5013(print)
Volume No.5, Issue Special 1 pp : 207-209 8 & 9 Jan 2016

The appropriate equations are mentioned in equation no. 1 & 2 Slope is assumed to be:
in this paper. All the calculations are in English units. Depth of water at 1 km = 34m (approx.)
Accordingly, unit conversions have been performed and the
discharge has been calculated.
MacDonald & Langridge-Monopolis (1984) & Washington
(2007) proposed empirical equations are Volume Eroded and
Breach Development Time. It is calculated from reservoir
volume (Vw) and maximum water depth (Hw).Parameters like
volume of embankment eroded during breach formation, based
on the product of the reservoir volume (Vw) and maximum
water depth (Hw) are very well computed by the MacDonald & Figure 1: Sketch of Breach
Langridge-Monopolis method.[ii]
Wetmore and Fread (1984) provide an alternative to the
MacDonald & Langridge-Monopolis (1984) and Froehlich
(2008) equations for breach peak discharge. [iii]
Washington State (2007) took the MacDonald & Langridge-
Monopolis method and adjusted it based upon whether the dam
is made of cohesion less or cohesive material.
However, The Froehlich (2008) method is dependent only on:
a) The volume of the reservoir,
b) Height of the breach and
c) The assumed breach side-slope.
This method also distinguishes between piping and overtopping
failure. A failure mode factor (Ko) is used.
Froehlich equation stands valid because with his consideration
dams with greater height tend to produce shorter failure times
Figure 2: Section of Breach
for a given reservoir volume.[iv], [v]
To calculate breach development time

Slope = 0.894473
Substituting values of Slope and Volume of water in equation
Where: (1) gives Breach development time in hours
Ko = Failure Mode Factor
Hb = Height of breach in feet
Vw = Reservoir volume stored in acre-feet
Again, substituting these values in Froehlich equation; gives
To calculate Discharge: the discharge through dam breach from equation no. (2)

Where:
Qp = Dam break peak discharge in cfs
Bavg = Average breach width in feet Assuming Breach width of 150 meters, discharge was
Hw = Maximum depth of water stored behind the breach in feet calculated as by using the
Tf = Breach development time in hours Froehlich empirical equation. As per this equation the breach
γ = Instantaneous flow reduction factor = 23.4 As/Bavg development time in hours was calculated as
(equivalent to ‘C’ in Wetmore and Fread (1984) .
As= Surface area of the reservoir in acres corresponding to Hw Same breach dimension was modelled in ANSYS Fluent
(CFD) software. Volume of Flow (VOF) model was used to
III. Results and Tables simulate the dam break.[vi], [vii], [viii]The discharge was
Manual Calculation for Discharge through dam breach using calculated as 81520cfs, i.e.2308.38 m3/s. Thus, the results for
Froehlich equation both are in close agreement with about 3% error which is
(All the dimensions in Fig. no. 1 are in meters, however acceptable for this type of simulation.
wherever necessary the values are converted to British units) Results for 2-D dam break analysis using ANSYS Fluent –
Dam Dimensions are as follows: VOF Model has been presented in figure 3[ix] [x].
Length = 1329 m
Total height = 42.7 m
Freeboard = (42.7-38) = 4.7 m
Total Volume = 30,500 km3
NCICE@2016 doi : 10.17950/ijer/v5i1/049 Page 208
 International Journal of Engineering Research ISSN:2319-6890(online),2347-5013(print)
Volume No.5, Issue Special 1 pp : 207-209 8 & 9 Jan 2016

Precautionary measures which can be taken are:

1) Monitoring: visual inspection which can be on site.
2) Dam logbook: a logbook in which activities (such as
maintenance and inspections) are recorded.
3) Dam safety review: at regular intervals, dam must
undergo a safety review by an engineer.[xiii]

IV. Conclusion
Time = 0 sec It is very difficult to estimate the cost of maintenance and
safety precautions for the given dam.The estimated cost of
rehabilitation of villages is Rs. 26,355 lacswhich is much
morethan the dam maintenance cost. Thusit can be said
thatprecaution is better than cure.

Acknowledgement
I hereby take opportunity to give my sincere thanks to Dr. S.
K. Ukarande and Prof. Seema A. Jagtap for their guidance
and constant encouragement and support in this paper. I truly
Time = 5 sec
appreciate the value and their esteemed guidance and
encouragement which would be remembered lifelong.

References
i. Office of the state engineer dam safety branch, February 10, 2010,
guidelines for dam breach analysis, state of colorado department of
natural resources division of water resources
ii. MacDonald, T.C., and Langridge-Monopolis. 1984, Breaching
Characteristics of Dam Failures, Journal of Hydraulic Engineering,
Vol. 110, No. 5,, pp. 567-586
Time = 10 sec iii. Wetmore, J. N. and Fread, D. L. (1984). The NWS Simplified Dam
Break Flood Forecasting Model forDesk-top and Hand-held
Figure 3: CFD Simulation Results for time 0,5 & 10 sec. Microcomputers. Federal Emergency Management Agency. 1984.
It was observed that, villages such as Phagne, Kale, Bramnoli, iv. Froehlich, D. C. 1995a, Peak Outflow from Breached Embankment
Kothume, Pavanangar, Yelse, Shivali, Bhodavali, Kadhade, are Dam, Journal of Water Resources Planning and Management, pp. 90-
97.
in very close proximity to the pavna dam and are highly prone v. Froehlich, D. C. San Antonio, Texas: s.n., 1995b, Embankment Dam
to the flood from the bam break. Breach Parameters Revisited, Water Resources Engineering,
Villages like Bour, Thugaon, Malvandi, Shivane, Oxarde, Proceedings of the 1995 Conference on Water Resources Engineering.
Pimpalkhule, Adhe are a bit far from the dam but still come pp. 887-891
vi. ANSYS CFX Introduction, ANSYS CFX Intro. Document for CFD.
under the flood affected area by the dam breach.[xi] Release 15.0, November 2013
About 9 villages (i.e: Phagne, Kale, Bramnoli, Kothume, vii. C. Biscarini, S. Di Francesco. 2010, CFD modelling approach for
Pavanangar, Yelse, Shivali, Bhodavali, Kadhade) can be dam break flow studies, Hydrol. Earth System Sciences, pp. 705-718.
completely under flood and might need rehabilitation. viii. Cameron T, Ackerman. Damfailure analysis using hec-ras and hec-
georas. Davis, CA: Hydraulic Engineer and Senior Technical
Table 1: Table showing Rehabilitation cost Hydraulic Engineer.
Name of Human No. of Rehabilitati ix. Chang, Tsang-Jung. 2011, Numerical simulation of shallow-water
Sr. dam break flows in open channels using smoothed particle
the Population Famalies on cost
No. hydrodynamics, Journal of Hydrology, pp. 78-90.
Village (Approx) (Approx.) (in lacs)
x. Xiong, Yi (Frank), 2011, A Dam Break Analysis Using HEC-RAS.
1 Phagne 594 100 1,500 xi. Irrigation Department, Pune Irrigation Circle, Pune - 411011
2 Kale 2,480 454 6,810 xii. Guidelines for Development and Implementation of Emergency Action
3 Bramnoli 648 189 2,835 Plan (EAP), May, 2006, for Dams, government of India central water
4 Kothume 1,102 354 5,310 commission dam safety organisation, New Delhi.
5 Yelse 1,041 182 2,730 xiii. The Regulation published in the Gazette of Dam Safety Act and
Regulation Government of Quebec, Canada
6 Shivali 1,289 206 3,090
7 Bhodavali 342 55 825
8 Kadhade 1,250 217 3,255
TOTAL 8,746 1,757 26,355

Thus, the total cost for rehabilitation of the villages is estimated

to be Rs. 26,355 lacs. This has been calculated considering 15
lac per family which includes land, construction, and infracture
cost.[xii]

NCICE@2016 doi : 10.17950/ijer/v5i1/049 Page 209