A Study Of Different Types Of Dam Construction

Introduction :
Dams are structures constructed across rivers to serve various purposes. Construction of dams has been in
practice since a very long time. With the progression of time, various types of dams were constructed based
on the location and prerequisite of society and designs were enhanced with the advancement of technology.
The types of dam commonly found are:-

 Earthen/Rockfill dams

 Concrete and Multi-arch dams

 Gravity Dams

 Buttress dams

 Concrete, steel, timber and composite material dams.

With all the benefits, constructions of dams may prompt to tremendous problems on the failure of the dam
structure. Dam failure is generally catastrophic and may occur due to many reasons such as piping through
the structure, overtopping, design error, heavy rainfall generated runoff, earthquake, etc. There are total 13
causes of failure listed in Hydrologic Engineering Centre (HEC) research document [1]. Whatever might be
the reason, the results are devastating. On failure of a structure the energy stored behind the dam can cause
a rapid and unexpected flood in downstream of the dam, resulting in loss of life and property. Middlebrooks
[2] specified scenarios for dam failure cases. Table 1 shows the causes of Earth dam failures due to various
reasons. As a dam break possess high hazards, dam break analysis is considered very essential. Dam break
analysis can be carried out by either Mathematical simulation using the computer or scaled physical
hydraulic model. According to Dam Safety Office, Water Resources Research laboratory Reclamation (1998)
[3] methods of analysis are grouped into four categories:-

(i) Physically Based Methods: Using erosion models based on principles of hydraulics, sediment transport
and soil mechanics, development of breach and resulting breach outflow are anticipated.

(ii) Parametric Models: Time to failure and ultimate breach geometry are assessed utilizing case studies and
then breach growth is simulated as a time dependent linear process and breach outflows are computed
using principles of hydraulics.

(iii) Predictor Equation: Using data of case studies, peak discharge is estimated from empirical equations and
a reasonable shape of outflow hydrograph is assumed.

(iv) Comparative Analysis: Breach parameters are determined by comparison of dam under consideration
and a dam that failed.
II. LITERATURE REVIEW:

Dam break has been a theme of concern and research for a long time. Dam break study incorporates
occurrence and proliferation of breach with time and analysis of the subsequent flood. Extensive research
has been done in the area of prediction of breach shape and its alteration with time. There is an immense
literature and case histories available about dam break modeling. Cristofano (19650 [4] considered the angle
of repose of given soil as dominant input for the estimation of the process of the breach erosion. Harris and
Wagner (1967) [5] predicted breach flows for a dam breach of parabolic shape while considering some
assumptions for breach parameters and sediment properties. Johnson and Illes (1976) [6] described failure
shapes of arch dams, gravity dams, and earthen dams.He explicated trapezoidal and few triangular breach
shapes particularly for earthen dams. Singh and Snorrason (1982) [7] studied 20 dam failures and deduced
the variation of breach width from 2 to 5 times the height of the dam. They observed that it will take 15
minutes to 1 hour for the complete failure of the dam and in the case of failure due to overtopping, the
maximum depth before failure ranged between 0.15 to 0.61 meters. MacDonald and Langridge-Monopolis
(1984) [8] introduced breach formation factor as the product of breach outflow volume and the depth of
water above the breach during failure. They analyzed 42 case studies and concluded that the breach side
slope could be assumed to be 1H: 2V in most of the cases, considering the breach shape to be triangular or
trapezoidal. Singh and Snorrason (1984) [9] analyzed 8 hypothetical breached dams and compared the
results of DAMBRK and HEC-1. They predicted peak outflows by varying breach parameters using both the
models. From the conclusion of their work, they showed that for large reservoirs the change in breach width
(Bw) produced large changes in the range of 35 to 87% in peak outflow in comparison of small reservoirs
that produced small changes in the range of 6 to 50 %. They also observed that NWS produced smoother
and reasonable flood stage profiles than those predicted by HEC. For steep slopes both the models
performed well but for mild slopes, HEC model predicted oscillating and erratic flood stages as HEC model is
unable to route flood waves in nonprismatic channels. Petra Check and Sadler (1984) [10] By changing the
breach parameters i.e. breach width and breach formation time they studied the sensitivity of discharge,
flood levels and flood arrival time. From the results of their analysis, they concluded that both parameters
have a reasonable impact for locations close to the dam whereas for locations well downstream of the dam,
both peak discharge and flooding levels are insensitive to changes in breach parameters while the timing of
flood wave peak can be modified by changes in breach formation time. Froehlich (1987) [11] analyzed
comprehensive case studies of real dam failures. From his analysis, he developed non dimensional prediction
equations for the estimation of the average breach width, formation time and average side-slope factor on
the basis of data obtained from case studies. By considering all the factors being equal, he also concluded
that breaches caused by overtopping are wider and also erode faster laterally in comparison of breaches
caused by other means.
Wurbs (1987) [12] performed numerous scenarios and concluded that breach simulation contains the
greatest uncertainty among all aspects of dam breach flood wave modeling. He also mentioned that with the
variation of reservoir size, the importance of different parameters also varies. He analyzed that in the case of
large reservoirs, the peak discharge occurs when breach attains maximum width and depth, therefore, it is
most critical to accurately predict breach geometry. But in the case of small reservoirs, peak discharge
occurs before the development of breach and hence, breach formation rate is considered to be a crucial
parameter in these cases.

Singh and Scarlatos (1988) [13] analyzed 52 case studies and documented about characteristics of breach
geometry and time of failure tendencies. They concluded that the ratio of top breach width to dam height
was considerably scattered and found that the ratio of top and bottom breach widths were in the range of
1.06 to 1.74 having an average of 1.29 with the standard deviation of 0.18. In most of the cases, side slopes
of the breach were inclined at 10-15º from vertical. They also deduced that most failure times were within 3
hours and 50% of the failures times were within 1.5 hours.

Von Thun and Gillette (1990) and Dewey and Gillette (1993) [14] They developed guidelines for the
estimation of breach parameters such as breach width at mid-height, side slopes and time to failure by
analyzing data from MacDonald and Langridge-Monopolis (1984) and Froehlich (1987). They suggested that
side slopes of breach should be 1:1 except for dams having cores of highly cohesive nature where slopes of
1:2 and 1:3 may be appropriate.

Y. Xu and L. M. Zhang, M.ASCE (2009) [15] studied earth and rockfill dam failure cases of 182 dams among
which 50% of dams were having the height above 15 meters i.e. large dams. They came up with a nonlinear
regression model by selecting 5 dam and reservoir control variables i.e. dam type, dam height, failure mode,
dam erodibility and reservoir shape coefficient and developed empirical relationships between five
breaching parameters (breach width, breach top width, average breach width, peak outflow rate and failure
time) and also evaluated importance of relativity among each control variables. And dam erodibility factor
was found to be most influencing breach parameter.

L.Y. Sidek et al. (2011) [16] Performed dam break modeling for a saddle dam for breach under Probable
Maximum Flood scenario and clear day scenario. From their analysis, they predicted dam breach using
Froehlich and MacDonald and Langridge-Monopolis predictor equation.

Rasif Razach (2014) [17] In his study he incorporated flow data and Digital Elevation Model as input in
HECRAS and created flood plain maps using ArcGIS by importing the output from HEC-RAS.