INVESTIGATION OF ARTIFICIAL CHANNELIZATION ON BRAHMAPUTRA RIVER BY NUMERICAL MODELING
A DISSERTATION Submitted in partial fulfillment of the requirements for the award of the degree of MASTER OF TECHNOLOGY in WATER RESOURCES DEVELOPMENT (CIVIL)
AFEWORK ASHAGRI s1RAL LIA R (2-)88t 14P C ACC• N3 ......
T• ROOR~~~ Jam'` L S~Jv X 4 X >
DEPARTMENT OF WATER RESOURCES DEVELOPMENT AND MANAGEMENT INDIAN INSTITUTE OF TECHNOLOGY ROORKEE ROORKEE -247 667 (INDIA) JUNE, 2012 INDIAN INSTITUTE OF TECHNOLOGY ROORKEE
CANDIDA IF 'S DECL21 k lTION
I hereby certify that the work which is being presented in the thesis entitled "INVESTIGATION OF ARTIFICIAL, CHANNELIZATION ON BRAHMAPUTRA RIVER BY NUMERICAL MODELING" in partial fulfillment of the requirements for the award of the Degree of MASTER OF TECHNOLOGY in Water Resources Development (Civil Engineering) and submitted in the Department of Water Resources Development and Management of the Indian Institute of Technology Roorkee, is an authentic record of my own work carried out during a period from July, 2010 to June, 2012 under the supervision of Dr. Nayan Sharma, Professor, Department of Water Resources Development and Management, Indian Institute of Technology Roorkee, India.
The matter presented in this thesis has not been submitted by me for the award of any other degree of this or any other Institute.
Place: Roorkee
Date: 41 ,L ~oI L- (AFEWORIC ASHAGRIE)
CERTIFICATE
This is to certify that the above statement made by the candidate is correct to the best of my knowledge.
(Dr. Nayan Sharma) Professor, Department of Water Resources Development and Management IIT Roorkee Roorkee-247667 India ABSTRACT
Brahmaputra river system is very complex because of its braided flow pattern and high sediment load. It is one of the most braided river in the world in which the width of the river extends up to 22 km. Constructing hydraulic structures like a bridge across the river along wide cross section will be difficult unless channelization works are provided.
The first part of the study is developing a mathematical model for the river by using HEC RAS 4.1.0. Sediment transport capacity of the river has been done by using different method of transportation functions like Ackers-White, Yang and Engelend-Hansen methods. The result shows that Yang's method of transportation function gives a better result for Brahmaputra River when it is compared with the other transportation functions. From these transportation functions Ackers-White method of transportation function gives high value of sediment transport capacity and high value of change in channel invert.
The second part of the study is to channelize the river by putting river training structures at selected nodal points. These nodal points are selected . based on the channel geometry and property in which channelization of the river can be possible. Some cross sections of the river are much braided in nature and it may not be economical to confine the flow over the whole reach of the river
Flow guide bunds are adopted as a river training structure to confine and guide the flow to a gentle channel. Especially these structures are constructed on the banks of the river at bridge site locations to guide and confine the flow. Other structures which are adopted as channelization work are closing of secondary channels by putting bed bars, permeable• spurs, porcupine screens, Jack Jetty etc.
The change of flow parameters have been compared before and after channelization of the river. There is an increase of velocity, shear stress, and sediment capacity at the cross section in which channelization is made' and up to a certain distance downstream of it. The water surface elevation also increases in the upstream sections up to a certain distance depending on the magnitude of the waterway and depth of channel.
Key Words: Numerical Model, Sediment Transport, Channelization, Hump/Bed Sill ACKNOWLEDGMENTS
Above all I want to thank the Almighty GOD for his eternal Love and helping me in my stay in India and through my life.
I would like to express my gratitude to my guide Dr. Nayan Sharma for his invaluable guidance, encouragement, and advice. I highly appreciate his constructive and valuable advices during my study especially on my research work. Without his support this work would not be realized. My gratitude goes to all professors in the department of Water Resources Development and Management for their help during my study in IIT Roorkee.
I would like to thank ITEC for granting me to pursue my study and Indian Institute of Technology, Roorkee for giving me the chance to study in India.
At last but not least, I would like to express my deepest thanks for my family supporting me through my life. TABLE OF CONTENTS ABSTRACT...... i ACKNOWLEDGMENTS...... ii TABLE OF CONTENTS ...... iii LISTOF FIGURES ...... vi LISTOF TABLES ...... ix 1. INTRODUCTION ...... 1 1.1. General ...... 1 1.2. Objectives ...... 3 1.3. Methodology ...... 3 1.4. Relevance of Study ...... 4 2. LITERATURE REVIEW ...... 2.1 Introduction ...... 5 2.2 Sediment Transport in Alluvial Rivers ...... 5 2.2.1 Stream Slope ...... 6 2.2.2 Stream Bed changes during the floods ...... 7 2.2.3 Degradation and Aggradations ...... 7 2.2 Numerical Modeling of rivers ...... 8 2.2.1 Introduction ...... 8 2.2.2 One-Dimensional and Multi-Dimensional Numerical Models ...... 9 2.2.3 Available Numerical Models for River Engineering ...... 10 2.4 HEC RAS, River Analysis System ...... 12 2.4.1 Steady Flow Water Surface Profiles ...... 13 2.4.2 Unsteady Flow Simulation ...... 14 2.4.3 Sediment Transport/Movable Boundary Computations ...... 14 2.5 Theoretical Basis for One-Dimensional Flow Calculation ...... 15 2.5.1 Equations for Basic Profile Calculations ...... 15 2.5.2 Cross Section Subdivission for Conveyance Calculation ...... 16 2.5.3 Composite Manning's n for the main channel ...... 17 2.5.4 Evaluation of Mean Kinetic Energy Head ...... 18 2.5.5 Friction Loss Evaluation ...... 19 2.5.6 Contraction and Expansion Loss Evaluation ...... 20
iii 2.5.7 Bed Roughnes Functions ...... 20 2.5.8 Sediment Transport Capacity ...... 20 2.5.9 Sediment Gradation ...... 20 2.5.10 Sediment Transport Functions ...... 21 2.6 Channelization of River ...... 25 2.6.1 Some Channelized Rivers ...... 25 2.7 Previous Studies on Brahmaputra River ...... 29 3. DESCRIPTION OF STUDY AREA ...... 30 3.1 Introduction ...... 30 3.2 Description of Study Area ...... 30 3.2 Longitudinal Section of Brahmaputra River ...... 33 4. NUMERICAL MODELING AND FLOW SIMULATION ...... 34 4.1 Introduction ...... :...... 34 4.2 Data Sources and Data Types ...... 35 4.2.1 Hydrological data ...... 35 4.2.2 Sediment Data ...... 3 5 4.2.3 Morphological data ...... 35 4.2.4 Satellite and DEM data ...... 36 4.3 Data Processing ...... 37 4.4 Model Development ...... 38 4.4.1 Data Requirements and input ...... 39 4.4.2 Energy Loss Coefficient ...... 47 4.5 Results and Discussions ...... :...... 47 4.5.1 Calibration of Bed Roughness ...... 47 4.5.2 Sediment Transport Analysis ...... 49 5. SIMULATION STUDIES FOR CHANNELIZATION OF BRAHMAPUTRA RIVER ...... 55 5.1 Introduction ...... 55 5.2 Types and Methods of Channelization ...... 55 5.3 Hydrological and Sediment Data ...... 58 5.4 Channelization at Different Nodal Points ...... 60 5.4.1 Introduction ...... 60 5.4.2 Channelization at Cross Section 57 ...... 61 5.4.3 Channelization at Cross Section 55 ...... 69
iv 5.4.4 Channelization at Cross Section 50 ...... 74 5.4.5 Channelization at Cross Section 45 ...... 76 5.4.6 Channelization near Tezpur ...... 80 5.4.7 Channelization at Cross Section 29 ...... 84 5.4.8 Cross Section 22 (Guwahati) ...... 87 5.4.9 Channelization at Downstream Cross Section of Guwahati ...... 89 5.4.10 Cross Section 9 (Jogighopa) ...... 91 6. NUMERICAL SIMULATION ON THE EFFECT OF SUBMERGED BED SILLS ON RIVER.... 92 6.1 Introduction ...... 92 6.2 Effect at the point of Submerged Bed Sill ...... 92 6.3 Upstream and Downstream Cross Section Effects ...... 93 7. SUMMARY AND CONCLUSIONS ...... 99 ANTEXES ...... 101 Appendix A: Downstream and Upstream Effects of cross section 57 ...... 101 Appendix B: Downstream Effects of Cross Section 55 ...... 105 Appendix C: Downstream Effects at CS 45 ...... 109. Appendix D: Downstream and Upstream Effect at CS 35.3125* ...... 112 Appendix E: Downstream Effect at Cross Section 29 ...... 115 Appendix F: Downstream Effects at Cross Section 21.6*, 21.4*, 21.2* and 21 ...... 118 Appendix H: Simulated and Observed Water Surface Elevation Using Different Method of Transportation...... 122 BIBLIOGRAPHY...... 124
V LIST OF FIGURES
Figure 2.1: Representation of Energy Equation terms ...... 16 Figure2.2: Mean kinetic energy ...... 18 Figure 2.3: Sketch of Yellow river Basin ...... 27 Figure 2.4: View of Lower Yellow River ...... 27 Figure 3.1: River Basin for the study area ...... 31 Figure 3.2: Map of India and study Area of Brahmaputra River ...... 33 Figure 3.3: Longitudinal Section of the study reach ...... 33 Figure 4.1: Satellite image of Brahmaputra River in 2008 ...... 36 Figure 4.2: DEM data for Brahmaputra River ...... 36 Figure 4.3: Catchment area of Brahmaputra River ...... 38 Figure 4.4: Flow series at Kobo ...... 40 Figure 4.5: Lateral Flow Series at Pandu ...... 40 Figure 4.6: Lateral Flow Series at Pandu ...... 40 Figure4.7: Rating Curve at Dhubri ...... 42 Figure 4.8: Representative Bed Gradation (Semi log) plot of the study reach ...... 44 Figure 4.9: Sediment Rating Curve at cross section 22 (Pandu) ...... 45 Figure 4.10: Geometric Profile of Brahmaputra River with interpolated cross section ...... 46 Figure 4.11: Observed and Simulated values at cross section 9 (Jogighopa) ...... 48 Figure 4.12: Observed and Simulated Values at cross section 22 (Pandu) ...... 48 Figure 4.13: Change in Invert (m) along the river starting from Cross section 2 (0.0)...... 52 Figure 4.14: Change in Invert (m) along the river starting from Cross section 2 (0.0) using Yang...... 52 Figure 4.15: Change in Invert (m) along the river starting from Cross section 2 (0.0) ...... 54 Figure 5.1: Sediment Rating Curve at Cross section 22 (Pandu) ...... 59 Figure 5.2: Sediment Rating Curve at cross. section 9 (Jogighopa) ...... 59 Figure 5.3: Surveyed width at different cross section of Brahmaputra River ...... :...... 60 Figure 5.4: Satellite images Cross section 57 ...... 61 Figure 5.5: Water surface elevation before and after Guide bund at July 22, 2004...... 62 Figure 5.6: Cross section Vs Velocity before and after construction of Guide bank ...... 64 Figure 5.7: Change in velocity through the simulation period at cross section 57...... 65 Figure 5.8: Shear stress time series at cross section 57 ...... 65 Figure 5.9: Mass bed Change at CS 57 before and after channelization ...... 66 Figure 5.10: Mass Capacity at cross section 57 before and after channelization ...... 66 Figure 5.1,1: Sediment Volume Capacity before channelization at CS 57...... 67 Figure 5.12: Sediment Volume Capacity at cross section 57 after guide banks...... 67 Figure 5.13:. Mass bed change before and after channelization at cross section 56.7777*...... 68 Figure5.14: Cross Section 55 ...... 69 Figure 5.15: Water surface Elevation change upstream of cross section 55...... 70 Figure 5.16: Velocity Change at upstream and downstream from cross section 55 ...... 70
VI Figure 5.17: Velocity time series at cross section 55 ...... 71 Figure 5.18: Shear stress at cross section 55 ...... 71 Figure 5.19: Mass Capacity at cross section 55 ...... 72 Figure 5.20: Sediment volume capacity at cross section 55 ...... 72 Figure 5.21: Sediment Volume capacity (m3/day) after channelization ...... 73 Figure5.22: Bed change at CS 55 ...... 73 Figure 5.23: Water surface elevation at Cross section 50 ...... 74 Figure 5.24: Water surface profile at CS 50 after channelization ...... 75 Figure 5.25: Velocity time series before and after channelization @ CS 50...... 75 Figure 5.26: Shear stress time series before and after channelization @ CS 50 ...... 76 Figure 5.27: Sediment Mass capacity before and after channelization at CS 50 ...... 76 Figure 5.28: Cross Section 45 ...... 77 Figure 5.29: Water surface Elevation Change ...... 78 Figure 5.30: Velocity change upstream and downstream section from CS 45 ...... 78 Figure 5.31: Velocity and Shear stress plots at CS 45 ...... 79 Figure 5.32: Mass capacity in tons/day at cross section 45 ...... 79 Figure 5.33: Seattleite images at Cross section 35.3125 ...... 80 Figure 5.34: Water surface elevation Change on 22 July, 2004 ...... 81 Figure 5.35: Change in Velocity at different cross section ...... 82 Figure 5.36: Velocity series at cross section35.3125* ...... 82 Figure 5.37: shear stress series at cross section35.3125* ...... 83 Figure 5.38: Mass Capacity in tons/day before and after channelization ...... 83 Figure 5.39: Satellite images at Cross section 29 ...... 84 Figure 5.40: Change in Water surface Elevation at different cross section ...... 85 Figure 5.41: Velocity change at different cross section in July 22, 2204 profile ...... 85 Figure 5.42: Velocity before and after channelization at cross section 29 ...... 86 Figure 5.43: Shear stress at cross section 29 before and after channelization ...... 86 Figure 5.44: Mass capacity at CS 29 before and after channelization ...... 86 Figure 5.45: Satellite images at Cross Section 22 ...... 87 Figure 5.46: Velocity at cross section 22 ...... 88 Figure 5.47: Shear stress at cross section 22 ...... 88 Figure 5.48: Mass Capacity at CS 22 ...... :...... :...... 88 Figure 5.49: Mass Capacity at CS 22 ...... :...... 89 Figure 5:50: Satellite images downstream of Guhawati ...... 90 Figure 5.51: Satellite images at Jogighopa ...... 91 Figure 5.52: Velocity, Shear stress. and Mass Capacity Plots at cross section 9(Jogighopa ...... 91 Figure 6.1: Velocity at cross section 22 before and after obstruction ...... 92 Figure 6.2: Cross section bed change at different times before obstruction ...... 93 Figure 6.3: Cross section Bed change at different times after obstruction ...... 93 Figure 6.4: Velocity at cross section 22.0526* before and after obstruction ...... 94 Figure 6.5: Velocity at cross section 22.1052* before and after obstruction ...... 94 Figure 6.6: Bed change before Obstruction at cross section 20.0526 ...... 95 vii Figure 6.7: Bed change after Obstruction at Cross section 20.0526 ...... 95 Figure 6.8: Velocity at cross section 21.9411 * before and after Obstruction ...... 95 Figure 6.9: Bed Change at cross section 21.9411* Before Obstruction ...... 96 Figure 6.10: Bed Change after Obstruction at cross section 21.9411* ...... 96 Figure 6.11: Bed Change before obstruction at Cross section 21.8823* ...... 97 Figure 6.12: bed Change After Obstruction at cross section 21.8823* ...... 97 Figure6.13: Optimum hump height ...... 98
viii LIST OF TABLES
Table 3-1: The Brahmaputra River: Country and Indian state-wise ...... 32 Table 4-1: Types of Data available ...... 35 Table 4-2: Drainage Area Ration at different points of the river cross section ...... 38 Table 4-3: Discharge and Stage Relation at Dhubri ...... 42 Table 4-4. Sediment gradation at Palasbari ...... 44 Table 4-5: Goodness of fit observed and simulated water surface elevations ...... 48 Table 4-6: Goodness of fit for water surface elevation ...... 49 Table 4-7: Goodness of fit at Pandu and Jogighopa ...... 50 Table 4-8: Change in invert Elevation at different time ...... 51 Table 4-9: Degradation at some cross section at different times ...... 52 Table 5-1: U/s and d/s length of the guide banks Gale's guide line ...... 58 Table 5-2: Water Surface Elevation and Velocity before and after Channelization upstream and downstream of cross section 57 ...... 63
ix 1. INTRODUCTION
101. General The study of reverian. system is required to know the pattern of the river flow and sediment transport. This will help to have some idea about the impact of the river in different scenarios. Some structures across the river may be constructed for irrigation, water supply, hydropower and other purposes. In addition to these studying the behavior of the river will give to awareness to the peoples to protect themselves and their properties and infrastructures from the risk of flooding.
River morphology has been a subject of great challenge to scientists and engineers who recognize that any effort with regard to river engineering must be based on a proper understanding of the morphological features involved and the responses to the imposed changes. River morphology is often presented from the geomorphic viewpoint. In the early stage, the regime concept originated from the study of stable alluvial canals, which with a mobile bed and earth banks, are non-scouring and non-silting over an operating cycle. A regime river is a system in dynamic equilibrium, or to be more precise, a system in quasi-equilibrium, for which the sediment transport is balanced by the sediment supply.
Analytical determination of the hydraulic geometry for river channels may be accomplished if the applicable physical relations are sufficient to describe the unknowns or degrees of freedom in channel morphology. The degrees of freedom of a river, in the broad sense, consist of the width, depth, channel slope, and bank slope. Whereas the channel depth can be determined from the e flow-resistance relation, that is the stage-discharge predictor for a given discharge, a physical relation that governs the stable width formation is not nearly as obvious. An important missing link that hindered rational effort by earlier researchers was the physical condition governing the width formation of alluvial channels. An alluvial river is unconstrained, in the long term, in developing its stable width to which the depth, slope, velocity, and flow resistance are closely related. Indeed, all boundaries of an alluvial river are free surfaces. The question of what mechanism under-lies the width formation must be answered before a rational determination of
1 channel geometry can be accomplished. Otherwise, river mechanics could only be at the unsatisfactory state of development.
River channel behavior often needs to be studied for its natural state and responses to human activities. Studies of river hydraulics, sediment transport, and river channel changes may be through physical modeling or mathematical modeling, or both. Physical modeling has been relied upon traditionally to obtain the essential design information. It nevertheless often involves large expenditures and is time consuming in model construction and experimentation. What limits the accuracy of physical modeling is the scale distortion which is almost unavoidable whenever it involves sedimentation. Mathematical modeling of erodible channels has been advanced with the progress in the physics of fluvial processes and computer techniques. Since the actual size of a river is employed in mathematical modeling, there is no scale distortion. The applicability and accuracy of a model depend on the physical foundation and numerical techniques employed. (Chang, 2008)
Individual alluvial channels are classically sub- divided into two single-thread patterns: straight and meandering, and a multi-thread or braided channel pattern (Leopold and Wolman, .1957). Meandering channels are generally distinguished from straight ones by having a higher sinuosity, P, defined as the ratio of channel length to valley length. Brahmaputra River has a braided channel pattern.
Some rivers are much braided and caries a lot of sediments and they are difficult to manage. Water resources and civil infrastructure across the river will be economical if they are constructed across the river in which the width of the flow is narrow. This can be possible by artificial channelization works along the river. Channelization works or river training structures can be used to protect for flood control, bank protection and navigation works and other purposes. Generally studying the behavior of the river system and conducting model studies is the most crucial thing to use the rivers efficiently and to minimize the risks from.flooding.
Mathematical modeling of the water and sediment has made much progress during the last decades. A number of more or less ready to use numerical models, have been developed. These models can be used, to the estimation of morphological process due to natural causes and/or
2 human activities. In this study a one dimensional mathematical model of HEC RAS 4.1 will be used to analyze the flow of Brahmaputra River. In addition the effect of artificial channelization at selected nodal points across Brahmaputra river like guide bunds and other river training structures have been analyzed.
Li Objectives Brahmaputra River is a wide river and much braided river which needs channelization works to confine the flow width for different purposes like navigation, flood control, bank protection as well as sediment control. The main objective of the study includes,
1. Developing a numerical model for the Brahmaputra river 2. The magnitude of erosion, deposition and transport of capacity of sediment will be computed for the selected cross sections. 3. Channelization of river by providing river training works like flow guide bunds, spurs, and others at 'specific cross sections of the river banks 4. Investigating the effects of artificial nodal points on the fluvial regime by constructing river training structures like guide bunds and secondary channel closure
L3. Methodology The methodology of the study can be sorted in to the following parts.
First, collection of hydrological data which includes flow rate, water surface elevation, sediment load data, temperature data, and morphological data of the river at different station of Brahmaputra River have been conducted. Generating and processing of hydrological data whenever it is required at specific station of river have been done.
Satellite Digital Elevation Model is downloaded from different sites like Shuttle Radar Topography Mission (SRTM). SRTM 90m data are a 3 arc-seconds (90m) resolution DEM data which can be downloaded freely from http://srtm.csi.cgiar.org/. This DEM data have to be enhanced by ERDAS IMAGINE to make the DEM more interpretable for watershed delineation. Watershed delineation has been done by Arc Hydro tool. In addition Satellite image data have been collected for analysis of the behavior of the river morphology and to integrate it with GIS and HEC GeoRas tools.
3 Finally HEC RAS 4.1.0 which is one dimensional computer based simulation method is used to analyze quasi unsteady flow of the river and to develop a model for Brahmaputra River.
L4, Relevance of Study Developing of the model for the river has great significance to know the flow parameters and sedimentation properties of the Brahmaputra river. Channelization of the river has been done in some rivers of the world. But it has not been done in Brahmaputra river. Knowing the effect of the behavior of flow after channelization will plays great role to put river training structures at specific points and to manage the braided pattern of the river.
4 2a LITERATURE REVIEW
2.1 Introduction A river carries water, sediment and solute from the drainage area to the sea and is thus of interest to hydraulic engineers, geomorphologist's and sedimentologists. This is important to engineers because water is used for a variety of purposes by humanity; water courses are used as navigation channels, and also erosion, transportation and deposition of sediment cause a number of problems in the river and in the catchment that must be solved pragmatically. The direct effect of transportation of sediment and water from the geologist's and geomorphologist's point of view is that the structure and form of the river and adjoining areas are continually changed due to erosion and sedimentation.
Since the dawn of civilization, mankind has used rivers for supporting and sustaining life. This has been done by harnessing and controlling rivers for the benefit of people. In doing so, the regime or stability of the river is invariably disturbed. In discussing these problems caused by disturbance in the stability of rivers, it is desirable to define what geomorphologists call a graded stream (Mackin 1948). A graded stream, poised stream, balanced stream or a stream in equilibrium is defined by Mackin as the one in which channel dimensions and slope are so adjusted over a period of time that it carries incoming sediment load and water without appreciable erosion or deposition (R.J.Garde, 2006).
An essential and well known feature of alluvial channels is that their morphological, flow resistance and sediment transport characteristics adjust in response to prevailing flow and alluvial conditions. Alluvial streams carry extremely varying discharge and sediment loads. The ratio of maximum to minimum discharge can attain values as high as 1000 or more in many streams (R.J Garde & K.J. Ranga Raju, 2000).
2.2 Sediment Transport in Alluvial Rivers Sediment transport related to the identification and mitigation of flood hazard on alluvial fans in and and semiarid environment is a current and critical concern of the engineering profession. In particular, estimating the length and maximum depth of deposition or erosion that occurs during a flow event when there is a change in the longitudinal slope of the channel is an important
5 problem. Deposition occurs when the slope changes from steep to mild and erosion occurs when the slope changes from mild to sleep. Once, a flood is over, or in a gradual time span, large changes on the river bed are observed with banks or piers eroded, while other locations get covered or aggraded. It might be impacted that when extremely large floods with limited sediment supplies and high sediment carrying capacities occur in rivers with erodible bed and bank materials, scour will continue to take place within the erosive capacity of the stream till it approaches the minimum/optimum value required to transport the available material (R.J Garde & K.J. Ranga Raju, 2000).
2.2.1 Stream Slope In general the longitudinal slope of a stream shows a continual decrease along its length. Examination of stream profiles would show that slope is greatest near the source, decreasing more or less regularly as the river flows its course. Such reduction in slopes corresponds to a longitudinal profile which is concave upwards ('(R.J Garde & K.J. Ranga Raju, 2000). Several factors are responsible for this. The reasons put forward by different scientists and engineers have been summarized in the following lines.
Firstly size of bed material being transported decreases in downstream direction due to abrasion.
Secondly, in humid regions, the discharge in a stream increases in the downstream direction due to inflow from the tributaries. Unless there is a corresponding increase in sediment inflow, the stream would necessarily flatten to the extent required by the increased sediment and water discharge (R.J Garde & K.J. Ranga Raju , 2000).
Thirdly, the sediment contribution of the upper region of a drainage basin is large compared to the run-off contribution to the stream flow. This means higher sediment contribution necessitating higher slope. While the lower region of the same discharge basin contributes smaller sediment quantity compared to its run-off discharge contribution signifying flatter slope requirement (R.J Garde & K.J. Ranga Raju, 2000).
Fourthly, on lower part of river sediments are usually finer and the streams are narrow with greater depth to width ratio leading to higher hydraulic efficiency requiring flatter slope (R.J Garde & K.J. Ranga Raju, 2000).
a 2.2.2 Stream Bed changes during the floods On several alluvial streams, the stream bed elevation was seen to raise during flood while the bed was lower after the receded. On few other streams exactly opposite happenings have been recorded. These changes can be very rapid for example on the Missourti River at Omaha, Nebraska (USA), the bed was found to be scouring at a rate of 0.3 m per minute during a flood (R.J Garde & K.J. Ranga Raju , 2000).
In the simplest form to understand the process of bed profile variation one has to assess the inflow out flow of sediment discharge in the reach under the consideration.
If the incoming amount or the rate of the sediment upstream of the reach is higher than outgoing from the reach downstream, it is obvious that the difference of the two quantities must have been dropped within the length of the reach. This process of rising of the bed level is called aggradation.
Conversely, if the incoming amount or the rate of the sediment upstream of the reach is lower than outgoing from the reach downstream, then the difference of the two quantities must have been fulfilled by picking up the bed materials from within the length of the reach. This process of lowering of the bed level is called Degradation.
2.2.3 Degradation and Aggradations If the rate at which sediment entering a given reach of the stream is less than that at which it is going out, the excess sediment will be picked up from the bed and banks, and there will be lowering of bed level unless the bed is non-erodible; this is known as degradation or retrogression. On the other hand, if the rate at which sediment enters a given reach of a stream is greater than the rate at which it goes out, the channel bed experiences deposition of sediment; this is known as aggradation. Aggradation or degradation occurs over large lengths and both are slow processes.
When alluvial streams are partially obstructed by hydraulic structures such as bridge piers, guide bunds, spurs or abutments, the local flow pattern around the structure is drastically changed causing high velocities and shear stresses in the vicinity of the structure causing local lowering
7 of the bed level. This is known as local scour. Local scour occurring around structures such as bridge piers can endanger their foundations causing bridge collapse. (R.J.Garde, 2006).
22 Numerical Modeling of rivers
2.2.1 Introduction Natural river channel changes are determined by the mutual interaction of the water flow with the erodible materials in the channel boundary. Knighton(1984) indicates that `The basic mechanical principles are well established but a complete analytical solution is still a long way ofd largely because natural streams represent the movement of a fluid solid mixture in boundaries that are themselves deformable. Even the motion of a single particle ' cannot be described adequately and the problem becomes more complex when the boundary material is cohesive'_ (B.prezedwojski, 1995).
Numerical models are established based on the conservation laws and mathematics (Zhang, 2006); it has to deal with many true physical and mathematical parameters. One has to understand all these parameters and make sure all the parameters prepared for the simulations are in correct range.
When a model applied to field study, one has to calibrate the model with field data. Because the resistance to the flow is represented by roughness coefficient which varies with properties of sediment, bed form, channel geometry, and vegetation etc; this information has to be characterized and fit to the model. Calibrating the model and identifying the mean or distribution of resistance to the flow is always necessary. Since the real world is very complicated, one normally would not have complete channel roughness information.
Numerical models approximate physical problems; it has however all the components to represent the physics to be simulated (Zhang, 2006). Numerical modeling of alluvial flow, sediment transport and morphological evolution started half a century ago and, to date, a variety of Numerical models have been developed and are in widespread use.
The ability to -make accurate calculation of fluvial flow, sediment transport, the associated morphological evolution processes and water quality is vital in a period when the concern over the river environment and the influence of human intervention is increasing. The interaction r between sediment and turbulent flow is of fundamental interest in the field of two-phase flow, and modeling the strongly coupled flow sediment morphology system provides a problem of considerable interest in computational fluid dynamics. Fluvial sediment transport process has been an increasingly important subject in the fields of water resources engineering, hydrology, geographical, geological and environmental sciences, and more fundamentally fluid dynamics.
The exposure of the alluvial systems to the natural and variable environment(Climatic, geological and social, etc) adds to the complexity of the process of sediment transport and the resulting morphological evolution of rivers(Cao and carling 2002).
2.2.2 One-Dimensional and Multi-Dimensional Numerical Models Numerous river engineering problems can be conveniently investigated by means of mathematical models. Mathematical models must properly describe the physical processes and provide a numerical solution to a system of differential equations that are solved together with suitable boundary conditions and empirical relationships that describe resistance to flow and turbulence.
The differential equations describing river mechanics problems are usually simplified forms of the equations of conservation' of mass and momentum, leading to a set of partial 'differential equations involving two independent variables (time and space or two spatial variables).
Since about 1970 a number of numerical models have been developed to solve problems of morphological changes in alluvial streams.
The ID numerical model is the most commonly used tool in Hydraulic Engineering for evaluating effects of flood waves in rivers.
One-dimensional mathematical models are quite useful in prediction of bed and water surface profiles, average depth, velocity and transport rate as a function of x and t. These have been used for solving problems such as
a) bed level variation during flood in lower reaches of the river; b) sedimentation upstream of a dam; c) degradation downstream of a dam;
0 d) modification of a river profile due to construction of embankments and execution of cutoffs; e) changes in river morphology due to addition or withdrawal of sediment or water; f) long-term evolution of river bed.
2D depth-integrated models have been applied to predict surface runoff and sediment-transport rates. A quasi-steady approach can often be used, although some 2D and 3D unsteady-flow models are available. A significant number of 2D and 3D codes are commercially available, and some are readily available in the public domain. The fast development of computers makes new numerical solutions possible to river engineering problems of increasing complexity.
In many cases in which the vertical variation in flow velocity and turbulence are of little interest, vertically averaged horizontal 2D models can be used.
The flow phenomena in natural rivers are three dimensional, especially those at or near a meander bend local expansion and contraction, or a hydraulic structure. Turbulence is an essentially three-dimensional phenomenon, and three-dimensional models are particularly useful for the simulation of turbulent heat and mass transport.
2.2.3 Available Numerical Models for River Engineering There is many developed software used to Model Rivers. Some of them are listed as follows:
1. DAMBRK: - is a flood forecasting model for dam failure made by the US National Weather Service. It consists of two parts: program for computing the outflow hydrograph from a reservoir where the dam is breaking and routing of the hydrograph downstream, by solving the Saint-Venant equations. 2. ISIS: - is made by Hydraulic Research Wallingford, in the UK It is similar in functionality to MIKEI1, with a graphical user interface and computational modules for one-dimensional steady and unsteady flow (Olsen, 2007). 3. RAM-2 computer programmes are readily available for conducting 2-D river hydraulics analysis in the horizontal plane (Thomas & McAnally 1985, U.S. Department of Transportation 1989).
10 4. TELEMAC-2D: is used to simulate free surface flows in two dimensions of horizontal space. At each point of the mesh, the program calculates the depth of water and the two velocity components. TELEMAC-2D solves the Saint-Venant equations using the finite- element or finite-volume method and a computation mesh of triangular elements. It can perform simulations in transients and permanent conditions. 5. CHEN'S MODEL: Chen (1973) formulated a model based on saint-venant's continuity and momentum equations of unsteady flow of sediment laden water. This model is capable of flood and sediment routing in a gradually varied flow channel. He used sediment load functions from Einstein's Bed load function as well as Toffaleti's function. Chen for the first time formulated a mathematical model that included sediment transport for generalized use. His works have proved to be a landmark in the f eld of open channel modeling for sediment laden flow (Chen, 1973). 6. DASS Model: Das (1975) developed multi stream and compound stream flow models by adopting the uncoupled solution procedures to rout water and sediment in non-uniform channels with the capability to simulate bed level changes. 7. FLUVIAL MODELS (1987 and 1084): Chang and Hill (1976) developed this model in 1976. The same equations of St. Venant are solved. In the case of aggradation, the deposition is made starting from the lowest point in horizontal layers. A four point implicit finite difference schemes with uncoupled solution procedure is used to solve the equations. Channel width adjustments are used to reflect lateral migration. Manning's equation is used to represent resistance to flow. He also developed FLUVIAL 11 Model in 1984 which employs a space-time model in which space-time model is represented by the discrete cross sections along the river reach and the time domain is represented by discrete time steps. The model uses the concept enunciated by Langbein and Leopold that the equilibrium channel represents a state of balance with bank erodibility or coefficient of bank erosion to predict the bank changes. Fluvial 11 is undoubtedly a promising model for channel changes prediction. However the adoption of empirical bank erodibility factor appears to have constrained its universal applicability and may require considerable calibration efforts. This model cannot be applicable for a river of multi-channel configuration S. GSTARS (Generalized Sediment Transport model for Alluvial River Simulation): The U.S. Bureau of Reclamation has developed a series of computer models (GSTAR) for the simulation and prediction of sediment transport, scour, and deposition processes in alluvial rivers and reservoirs. GSTARS, GSTARS 2.0/2.1 and GSTARS3 are based on the stream tube concept using one-dimensional approach along stream tubes to obtain a semi-two-dimensional variation of the hydraulic conditions in rivers and reservoirs. 9. MIKE 11 and MIKE 21 C: These river modeling software are developed by DHI-Water and Environment, which is a research and consulting organization developing and applying advanced methods and technologies within hydraulic and water resources engineering. MIKE 11 is a one dimensional hydrodynamic software package including a full solution of the St. Venant equation, plus many process modules for advection-dispersion, water quality and ecology, sediment transport, rainfall-runoff, flood forecasting, real-time operations, and dam break modeling. MIKE 21C is a two dimensional river hydraulics and morphology model..It is one of the most well established tools for simulating the development in the river bed and channel. plan form caused by changes in the hydraulic regime. Simulated processes include bank erosion, scouring and shoaling brought about by activities such as construction and dredging or seasonal fluctuations in flows.
24 HEC RAS9 River Analysis System HEC is an abbreviation for Hydrologic Engineering Center. It. is a part of US Army Corps of Engineers. Over the years, the organization has made several computer programs for water flow problems, named HEC1, HEC2, HEC3, etc. HEC2 compute the water surface profile for a steady water flow in a natural river in one dimension. Now the latest version which was released in 2010 is version 4.1.0.
It is an integrated system of software, designed for interactive use in a multi-tasking, multi-user network environment. The system is comprised of a graphical user interface (GUI), separate hydraulic analysis component, data storage and management capabilities, graphics and reporting facilities.
12 The HEC-RAS contains four one-dimensional river analysis components for:
Steady flow water surface profile computations- > Unsteady flow simulations Movable boundary sediment transport computations and Water Quality
24J Steady Flow Water Surface Profiles. This component of the modeling system is intended for calculating water surface profiles for steady gradually varied flow The system can handle a full network of channels, a dendritic system or a single river reach The steady flow component is capable of modeling subcritical, supercritical, and mixed flow regime water surface profiles.
The basic computational procedure is based on the solution of the one-dimensional energy equation. Energy losses are evaluated by friction (Manning's Equation) and contraction/expansion (coefficient multiplied by the change in velocity head). The momentum equation is utilized in situations where the water surface profile is rapidly varied. These situations include mixed flow regime calculations (i.e., hydraulic jumps), hydraulics of bridges, and evaluating profiles at river confluences (stream junctions).
The effects of various obstructions such as bridges, culverts, dams, weirs, and other structures in the flood plain may be considered in the computations. The steady flow system is designed for application in flood plain management and flood insurance studies to evaluate floodway encroachments. Also, capabilities are available for assessing the change in water surface profiles due to channel modifications and levees.
Special features of the steady flow component include:
Multiple plan analysis Multiple profile computations > Multiple bridge and/or culvert opening analysis Split flow optimization Stable channel design.and analysis.
. 13 2.4.2 Unsteady Flow Simulation. This component of the HEC-RAS modeling system is capable of simulating one-dimensional unsteady flow through a full network of open channels. The unsteady flow equation solver was adapted from Dr. Robert L. Barkau's LTNET model (BARKAU, 1992 and HEC, 1997). The unsteady flow component was developed primarily for subcritical flow regime calculations. However the present model can now perform mixed flow regime (subcritical, supercritical, hydraulic jumps, and draw downs) calculations in the unsteady flow computations module.
The hydraulic calculations for cross-sections, bridges, culverts, and other hydraulic structures that were developed for the steady flow component were incorporated in the unsteady flow module.
Special features of the unsteady flow component include:
> Dam break analysis > Levee breaching and overtopping Pumping stations Navigation dam operations Pressurized pipe systems
204.3 Sediment Transport/Movable Boundary Computations This component of the modeling system is intended for the simulation of one-dimensional sediment transport/movable boundary calculations resulting from scour and deposition over moderate time periods (typically years, although applications to single flood events are possible).
The sediment transport potential is compared by grain size fraction, thereby allowing the simulation of hydraulic sorting and armoring. Major features include the ability to model a full network of streams, channel dredging, various levee and encroachment alternatives, and the use of several different equations for the computation of sediment transport. a] The model is designed to simulate long term trends of scour and deposition in a stream channel that might result from modifying the frequency and duration of the water discharge and stage, or modifying the channel geometry. The system can be used to evaluate deposition in reservoirs, design channel contractions required to maintain navigation depths, predict the influence of
14 dredging on the rate of deposition, estimate maximum possible scour during large flood events, and evaluate sedimentation in fixed channels.
205 Theoretical Basis for One-Dimensional Flow Calculation HEC-RAS is currently capable of performing one-dimensional water surface profile calculations for steady gradually varied flow in natural or constructed channels. Subcritical, supercritical and mixed flow regime water surface profiles can be calculated
205i Equations for Basic Profile Calculations Water surface profiles are computed from one cross section to the next by solving the energy equation with an iterative procedure called the standard step method. The energy equation is written as follows;
Y2 +Z2 + a2VZ Yl +ZI - alV1 °}-he (20l) ig = 2g Llg~, Where: Y1, Y2 = depth of water at cross sections ( G,4TRA 2J 881L '~ ACCMai.i...... • -4p Z1, Z2 = elevations of the main Channel inverts flate::.::iiiiiiiii...i.•
V1, V2 = average velocities (total discharge/total flow area) fi D ~
al , a2 = Velocity weighting coefficients
g = gravitational acceleration
he = energy head loss.
A diagram showing the terms of the energy equation is shown in the following figure,
15 -_Energy Grade Line 2g he
Water Surface
Y2 29
Bottom Y,
z2 Z,
Figure 2.1; Representation of Energy Equation terms The energy loss (he) between two cross sections is comprised of friction losses and contraction or expansion losses. The equation for the energy loss is as follows
he = LSf + C a2V2 I (2.2) I 2g 2g
Where: L = the weighted reach length and Sf is the representative friction slope between two sections and C is the expansion or contraction loss coefficent.
The distance weighted reach length, L, is calculated as
L __ L1obQ1ob+LchQch+LrobQrob (2.3) Q1ob+Qch+Qrob
Where: L 0b, Lrh, L ob = cross section lengths specified. for flow in the left overbank, main channel, and right over bank, respectively.
Qiob + Qch + Qrob = arithmetic average of the flows between sections for the left overbank, main channel, and right overbank, respectively.
205,2 Cross Section Subdivission for Conveyance Calculation The determination of total conveyance and the velocity coefficient for a cross section requires that flow be subdivided in to units for which the velocity is uniformly distributed. The approach used in HEC RAS is to subdivide flow in the overbank areas using the input cross section n break points (locations where n-values change) as the basis for. subdivision from the following form of Manning's equation.
16 Q. =KSf'Z (2.4)
K= n AR2/3 (2.5)
Where: K = conveyance for subdivision
n = Manning's roughness coefficient for subdivision
A = flow area for subdivision
R = hydraulic radius for subdivision (Area/wetted perimeter)
The program sums up all.the incremental conveyance in the overbanks to obtain a conveyance for the left overbank and the right overbank. The main channel conveyance for the cross section is obtained by summing the three subdivision conveyances (left, channel, and right).
2.5.3 Composite Manning's n for the main channel Flow in the main channel is not subdivided, except when the roughness coefficient is changed within the channel area. IIEC RAS tests the applicability of subdivision of roughness within the main channel portion of a cross section, and if it is not applicable, the program will compute a single composite n values for the entire main channel. The program determines if the main channel portion of the cross section can be subdivided or if a composite main channel n value will be utilized.
For the determination of n, the main channel is divided in to N parts and, each with a known wetted perimeter pi and roughness coefficient n;.
a/a " (piaix.$)l nc = (2.6) i=1 p J
Where: n composite or equivalent coefficient of roughness
P = wetted perimeter of entire main channel
P = wetted perimeter of subdivision I
n; = coefficient of roughness fr subdivision
17 The computed composite n,_ should be checked for reasonableness.
205.4 Evaluation of Mean Kinetic Energy Head Because the HEC-RAs software is a one dimensional water surface profiles program, only a single water surface and therefore a single mean energy are computed at each cross section. For a given water surface elevation, the mean energy is obtained by computing a flow weighted energy from the three subsections of a cross section (left overbank, main channel, and right overbank).
f,r2 V2 V2
Figure 2.2: Mean kinetic energy Where: V1 = mean velocity for subarea I
V2 = mean velocity for subarea 2
To compute the mean kinetic energy it is necessary to obtain the velocity head weighting coefficient alpha. Alpha is calculated as follows:
Mean kinetic Head = Discharge – Weighted Velocity Head
z z V2 Qlvs a— = Q2 28 (2.7) 29 Qi +Qz
z z 291QI8 Qz z9 3 a = (2.8) (Qi+Q2)7,
a Q1V12+Q2V22 = 2 (2.9) (Q1+Q2 )V
In general:
18 a _ (QlVj2+Q2V22+...+QNVN2) QV2 (2.10)
The velocity coefficient a is computed based on the conveyance in the three flow elements: left overbank, right overbank, and channel. It can be written in terms of conveyance and area as in the following equations:
bKZO eh+K (At)2f Zob a _ LA [o 6 Ach Ao6J K3 (2.11) t
Where: At = total flow area of cross section, Aiob, Ach, Arob = flow areas of left overbank, main channel and right overbank respectively, Kt = total conveyance corss section and K10b, Kch, '(rob = conveyance of left overbank, main channel and right overbank respectively.
2.5.5 Friction Loss Evaluation Friction loss is evaluated in HEC-RAS as the product of Sf and L, where Sf is the representative friction reach and L is defined by the equation (2.3) above. The friction slope (slope of the energy grade line) at each cross section is computed from Manning's equation as follows.
(2)2 S f = Q (2.12)
Alternative expressions for the representative reach friction slope (S f) in HEC-RAS are:
Average conveyance Equation
2 (2.13) S f (K1+K2)z
Average Friction Slope Equation
Sf = Sf1+Sf2 (2.14)
Geometric Mean Friction Slope Equation
Sf = Sfi + S2 (2.15)
Harmonic Mean Friction Slope Equation
19 S = Z(SfiXSf2) /2.16l f Sfz+Sf2 l i
2.5.6 Contraction and Expansion Loss Evaluation Contraction and expansion losses in HEC-RAS are evaluated by the following equation:
la,2 hCe = c v1 —a2V2` (2.17) 2g 2g
Where: C = the contraction or expansion coefficient.
The program assumes that a contraction is occurring whenever the velocity head downstream is greater than the velocity head upstream. Likewise, when the velocity head upstream is greater than the velocity head downstream, the program assumes that a flow expansion is occurring:
20507 Bed Roughnes Functions Because manning's n values are typically used in HEC-RAS, the uniform flow feature assumes for the use of a number of different roughness equations to solve for n. HEC-RAS allows the user to apply any of these equations at any area within a cross section; however the applicability of each equation should be noted prior to selection. Manning equation method, one n value or a single n vahies is prescribed across the cross section and then the manning's equation is used to solve for the desired parameter.
2.5.8 Sediment Transport Capacity Sediment transport capacity function in NEC RAS has the capability of predicting transport capacity for non-cohesive sediment at one or more cross section based on existing hydraulic parameters and known bed sediment properties. It does not take in to account, sediment inflow, erosion, or deposition in the computations. Classically, the sediment transport capacity is comprised of both bed load and suspended load, both of which can be accounted for in the various sediment discharge rating curves, which help to understand and predict the fluvial process found in natural rivers and streams.
2.5.9 Sediment Gradation Sediment transport rates are computed for the prescribed hydraulic and sediment parameters for each representative grain size. Transport capacity is determined for each grain size as if that
20 particular grain size made up of 100% of the bed material. The transport capacity for that size group is then multiplied by the fraction of the total sediment that size represents. The fractional transport capacities for all sizes are summed for the total sediment transport capacity.
gs = Di1 9SA (2.18)
Where: gs= total sediment transport
gs; = Sediment transport for size class i,
pi = Fraction of size class I in the sediment, and n = number of size classes represented in the gradation.
Because different transport functions were developed differently with a wide range of independent variables, HEC-RAS gives the user the option to select how depth and width are to be computed. The HEC-6 method converts everything to an effective depth and width. However, many of the sediment transport functions were developed using hydraulic radius and top width, or an average depth and top width. For irregular cross section shapes, HEC-RAS uses the effective depth/effective width or hydraulic radius/top width as the default. Also available for use is the hydraulic depth, which is used to represent the average depth and is simply'the total area of the section by the top width.
23510 Sediment Transport Functions Because different transport functions were developed under different conditions, a wide range of results can be expected from one function to the other. Therefore it is important to verify the accuracy of sediment prediction to an appreciable amount of measured data from either the study stream or a stream with similar characteristics. It is very important to understand the development of the functions in order to be confident of its applicability to a given stream.
Typically sediment transport functions predict rate of sediment transport from a given set of steady-state hydraulic parameters and sediment properties. Some functions compute bed-load transport, and some compute bed-material load, which is the total load minus the wash load (total transport of particles found in the bed). In sand bed streams with high transport rates, it is common for the suspended load to be orders of magnitude higher than that found in gravel-bed
21 or cobbled streams. It is therefore important to use a transport predictor that includes suspended sediment for such a case. The following sediment transport functions which are also available in HEC-RAS;
> Ackers-White(1973) > Engelend-Hansen (1967) > Laursen (1958) Meyer-Peter Muller (1948) > Toffaleti (1969) > Yang(1973)
These functions were selected based on their validity and collective range of applicability. All of these functions, except for Meyer-Peter Muller (1948), are compared extensively by Yang and Wan (1991) over a wide range of sediment and hydraulic conditions. Results varied, depending on the conditions applied. The Meyer-Peter Muller (1948) and the bed load portion of Toffaleti (1981). They concluded that Toffaleti (1969) bed-load procedure was sufficiently accurate for their test stream, whereby, Meyer-Peter Muller (1948) was not useful for sand-bed channels at or near incipient. A short description of three main sediment transport predictors is summarized below (Brunner, 2002; Karamisheva, 2006).
Ackers-White (1973)
The Ackers-White transport function is a total load function developed under the assumption that fine sediment transport is best related to the turbulent fluctuations in the water column and coarse sediment transport is best related to the net grain shear with mean velocity used as the representative variable. The transport variable was developed in terms of particle size, mobility and transport.
A dimensionless size parameter is used to distinguish between the fine, transitionary, and coarse sediment sizes. Under typical conditions, fine sediments are silts less than 0.04mm, and coarse sediments are sands greater than 205mm. Since the relationships developed by Ackers-White are applicable only to non-cohesive sands greater than 0.04mm, only transitionary and coarse
22 sediments apply. Original experiments were conducted with coarse grains up to 4mm: however the applicability range was extended to 7mm.
This function is based on over 1000 flume experiments using uniform or near-uniform sediments with flume depths up to 0. 4mm. a range of bed configurations was used, including plane, rippled, and dune forms, however the equations do not apply to upper phase transport (e.g. anti- dunes) with Froude numbers in excess of 0.8. The general transport equation for the Ackers- White function for a single grain size is represented by;
X =D(u}n and Ggr = C(Ar -1) (2.19)
Where: X = sediment concentration in parts per part
Gg, = Sediment transport parameter;
s = Specific Gravity of sediments
ds = Mean Particle parameter,
D= Effective Depth
u = Shear velocity, V = average velocity,
n = Transition Exponent, depending on sediment size; C = coefficient, Fg, = Sediment mobility parameter, A = Critical sediment mobility Parameter.
A hiding adjustment factor was developed for the Ackers-White method by Profitt and Sutherland (1983), and is included in RAS as an option. The hiding factor is an adjustment to include the effects of a masking of the fluid properties felt by smaller particles due to shielding by larger particles. This is typically a factor when the gradation has a relatively large range of particle size and would tend to reduce the rate of sediment transport in smaller grade classes (Brunner, 2002).
23 Engelend-Hansen (1967)
The Engelend-Hansen function is a total load predictor which gives adequate results for sandy rivers with substantial suspended load. It is based on flume data with sediment sizes between 0.019 and 0.93mm. It has been extensively tested, and found to be fairly consistent with field data. The general transport equation for the Engelend-Hansen function is represented by:
3/2 2 d so R ° gs = 0,05 V g(IS1) (2.20) kY5—Y)dso
Where: gs= unit sediment transport; T =unit weight of water, Ys = Unit weight of solid particles;
V = average channel velocity, T. = Bed level shear stress; d50 = particle size of which 50% is smaller.
Yang (1973)
Yang's method (1973) is developed under the premise that unit stream power is the dominant factor in the determination of total sediment concentration. The research is supported by data obtained in both flume experiments and field data under a wide range conditions found in alluvial channels. Principally, the sediment size range is between 0.062 and 7.0mm with total sediment concentration ranging from 10 ppm to 585, 000 ppm. Channel widths range from 0.44 to 1746 ft, depths from 0.037 to 49.4 ft, water temperature from 00 to 34.3°Celcius, average channel velocity from 0.75 to 6.45 fps, and slopes from 0.000043 to 0.029 (Yang and Wan, 1991).
Yang (1984) extended the applicability of his function to include gravel sized sediments. The general transport equations for sand and gravel using the Yang function for a single grain size is represented by: (Garde and Raju, 2000, Brunner, 2002)
log Ct = 5.435 — 0.2861ogd'n — 0.457log l + (1.799 — 0.409 flog`- — 0.314 log W*) Iog —s -- (2.21)
For d 24 log Ct = 6.681— 0.633 logdm — 4.8161og w + (2.784 — 0.305 log — 0.282 log) log (mss — Wr~v (2.22) For gravel d >= dm = 2mm Where: Cc = Total sediment concentration, co = particle fall velocity, dm = Median particle parameter v = kinematic viscosity, u' = Shear velocity, V = average channel velocity, S = Energy gradient. 26 Channelizatiwi of River The Brahmaputra River displays a braided pattern in plain view, and short term channel migration is quite drastic, with rates of movement as high as 2,600 ft. a year being common. The rate of rise and fall of the river, the number and position of major channels active during flood, the formation and movement of large bed forms, cohesion and variability in composition of bank material, and intensity of bank slumping are some of the factors responsible for controlling the bank line configuration and movement.. The most significant bank line modifications take place during falling-river stage, when excess sediment is deposited as bars within the channel, causing a change in local flow direction and migration of the thalweg (COLEMAN, 1968). 2.6.1 Some Channeled Rivers Channelization works have been done for some specific rivers in the world such as Mississippi river in USA, Rhone River in Europe, Yellow River in china, and also other rivers. This channelization works are done to affect the width, depth and other flow characteristics of the flow different purposes like for navigation, flood control, urbanization, and other tasks. Some of the studies on channelization of rivers are explained below The Yellow River in Northern and Northeastern China is the country's second largest river, with a drainage area of about 795,000 km2. It originates from the mountains of the Qinghai Plateau in Western China and flows 5,464 Ian to the Bohai Sea off the eastern coast of China. 25 The river training works constructed on the Lower Yellow River may be classified into two types according to the location of the structures (Xu et al. 1987; Zhang and Xie 1993; Hu et al. 1998), as shown in Figure 2.4 and 2.5 below. The first types are levee protection works, which are commonly constructed at vulnerable spots along and in connection with the levees. Their primary function is to protect the levee from scouring, and their secondary function is to regulate the flow. The second type are flow guide works, which are constructed along the banks of floodplains. Their primary function is to guide the flow through the planned channel alignment, and their secondary function is to protect the floodplains from scouring. To effectively control the flood flows in the planned channel alignment, a river reach usually includes a number of levee protection works and flow guide works (Baosheng Wu et al, 2005). The Lower Yellow River is usually divided into three geomorphically distinct reaches (Qian et al. 1993), as shown in Fig. 2.4. The upper reach, from Tiexie to Gaocun, has a typical braided channel pattern with a sinuosity as low as 1.15. It extends a length of 299 km and has a main channel width varying between 1 and 3.5 km. The flood plains on both sides along the channel are very wide. The distance between the levees is 5-20 km. The channel bed slope varies gradually from 0.000265 at the upstream to 0.000172 downstream. The channel depth is only about 2 m for flows under bank-full conditions. The lower reach, from Taochengpu to Lijin, with a length of 300 km, is a well-defined and stable meandering river reach with an average sinuosity of 1.21. The average channel slope is 0.0001 and the distance between levees is 0.45-5 km. The main channel has a narrow and deep section as compared to the braided reach, with a width 300-800 m. The river reach lying between Gaocun and Taochengpu, with a length of 165 km and an average sinuosity of 1.33, has a transitional channel pattern from braided to meandering. The average channel slope is 0.000115. The distance between levees is 1.4-8.51km, and the width of the main channel is 0.5-1.6 km. M Figure 2.3: Sketch of Yellow river Basin / (b) Figure 2.4: View of Lower Yellow River Human modifications to Mississippi river were introduced largely after 1920. Between 1929 and 1942, the river was channelized and shortened 245 km to improve and maintain navigation. As the result of shortening, 15 meander bends were cut-off and isolated from the main-stem channel. The river was shortened an additional 88 km between 1939 and 1955 by chute cutoffs. After their formation, cutoffs acted as sediment storage locations, essentially removing channel and point bar sediments from the active sediment budget of the main-stem channel (Kesel*, 2003). The overall influence of modifications on the sediment regime includes a major decrease in sediment input both from major tributaries and channel and flood plain sources. The most significant change has been a greater than 90% reduction in bank caving sediments as the result of revetment construction. Based on bank caving data from Smith (1963), most of this decrease occurred after 1941. A sizeable reduction in sediment input from the Missouri and Arkansas 27 rivers occurred after 1950 as the result of dam and reservoir construction (Keown et al., 1986; Kesel, 1988, 1989). Further reductions in the amount of sediment reaches the mouth of the river results from increased storage capacity. The construction of dike fields starting in 1955 promoted the accretion of bed load sediments, largely sand, creating or enlarging channels bars. By the mid-1970s, the amount of sediment, largely bed load, trapped in these dike fields (Mississippi River Commission (MRC), 1987) exceeded the volume of sediment stored on unregulated active bar locations. Channel segments isolated during the cutoff program in the 1930s and 1940s removed point bar and channel sediments from the sediment budget. As a result of these modifications, it may be expected that a negative balance would be produced in the sediment budget and the amount of sediment reaching the Gulf of Mexico would be greatly reduced, particularly the bed load portion. Prior to major modifications in the 1930s, the Lower Mississippi River was a classic meandering river that was aggrading its channel throughout much of its length. This aggradation is reflected in the growth of channel bars and the increase in thaiweg elevation prior to 1935. An estimate of the average annual sediment load reaching the Gulf of Mexico at that time included a suspended load of 270.106 m3/year and a bed load that may have been as much as 130.106 m3/ year (Kesel et al., 1992). The components of the Lower Mississippi. River' sediment regime include both sediment input sources and short- and long-term storage locations within the fluvial system. These data indicate that the river had two distinct sediment regimes which reflected the geomorphological subdivisions noted previously. As much as two-thirds of the sediment load transported by the river from Cairo to Red River Landing was generated by bank caving as the river meandered on the flood plain. Within this segment, approximately twice as much sediment was stored as short-term storage within the channel, mostly in river bars, as was stored as longer term overbank storage on the flood plain (Kesel et al., 1992). The growth of channel bars during the period was largely a function of bank caving. 28 2.7 Previous Studies on Brahmaputra River The Geological Survey of India has periodically (e.g. 1981) carried out geomorphological study of the Brahmaputra basin. The fluvial and morphology Brahmaputra river in Assam has been studied by J.N Sarma (2005) which includes the study of Geomorphology of the flood plain, geometry of the meanders, morphology of the river bed, bars, aggradations and degradation of bed and etc. Delft Hydraulics and DIII (1996a—c), Delft Hydraulics and Leeds University (1996) have provided the quantitative information about form and process within the Jamuna, Ganga and Meghna Rivers in Bangladesh. Channel braiding and stability of the Brahmaputra River, Bangladesh, since 1967 was studied using the help of GIS and Remote Sensing analysis by JALUX Inc., JAL Building, 2-4-11, Higashi-Shinagawa, Shinagawa-ku, Tokyo 140-0002, Japan(2006). 29 3. DESCRIPTION OF STUDY AREA 3J. Introduction Brahmaputra River is one of the major rivers in the Asia as well as in the world covering a drainage basin of 580,000 km2, extending from 82°E to 970 50' E longitudes and 25° 10' to 31° 30' N latitudes. It flows through Tibet (China), India, Bangladesh and Bhutan with a total drainage area of 580,000km2 at the Goalundo in Bangladesh. In India the Brahmaputra flows through the states of Arunachal Pradesh for 278km, mostly across the Himalayas, where it is called Dihang or Siang River. The Dihang emerges on the plains at Pasighat (elevation 155m). Near Kobo in Assam, 52 km downstream from Pasighat, the Dihang is joined by two larger rivers — the Lohit and Dibang, and from here the river is known as Brahmaputra River. The Brahmaputra river flows for about 670 km through the state of Assam along the Assam Valley and within Assam the Brahmaputra receives 103 tributaries which are 65 on the right (north) bank and 38 on the left (south) bank. Some of the larger tributaries are the Subansiri, the Jia Bharali, the Manas, and the Sonkosh on the right bank and the Buhri Dihing, the Dhansiri and the Kopili on the left bank. The Brahmaputra valley and its adjoining hill ranges are seismically unstable (Sarma, 2005). 3,2 Description of Study Area Before entering India, the river flows in a series of big cascades as it rounds the Namcha-Barwa peak. The river forms almost trough receiving the flows of its tributaries both from North and South. The river, with its Tibetan name Tsangpo in the uppermost reach, flows through southern Tibet for about 1,625 km eastward and parallel to tributaries, viz., the Nau Chhu, the Tsa Chhu, the Men Chhu, the Charta Tsangpo, the Raga Tsangpo, the Tong Chhu, the Shang Chhu, the Gya Chhu, the Giamda Chhu, the Po Tsangpo and the Chindru Chhu and the right bank tributaries, viz., the Kubi, the Kyang, the Sakya Trom Chhu, the Rhe Chhu, the Rang Chhu, the Nyang Chhu, the Yarlang Chhu, and the Trulung Chhu join the river along its uppermost reach. At the extreme eastern end of its course in Tibet the Tsangpo suddenly enters a deep narrow gorge at Pe, where in the gorge section the river has a gradient ranging from about 4.3 to 16.8 n /km. 30 The river enters in India near Tuning in Arunachal Pradesh. After travelling for a distance of 278 km up to Kobo, it meets with two rivers the Dibang and the Lohit in Assam near Kobo. Below this confluence point, the river is known by the name of the Brahmaputra. It passes through Assam into Bangladesh and at last it meets with the Ganga near Goalundo in Bangladesh before joining the Bay of Bengal. Its total length is 2,880 kin comprising of 1,625 km in Tibet, 918 km in India and 337 km in Bangladesh. It is also one of the most braided rivers in the world with width variation from 1.2 km at Pandu near Guwahati to about 18.13 km near Gumi few km distances downstream to this point. The river basin of the Brahmaputra is bounded on the north by the Kailas and Nyen- Chen- Tanghla ranges of mountains; on the east by the Salween river basin and the Patkai range running along the Indo-Myanmar border; on the south by the Nepal Himalayas, the Naga and Barail ranges and the Meghalaya Plateau; and on the west by the Ganga river basin. Legend DEAN-®mptimg VALUE ® 0-446 ® 446.0000001 - 1,118 1,118.000001-1,833 ■ 1,833.000001-2.580 ■ Z580.000001-3,357 ■ $357.000001-4,075 ® 4,075.000001-4,664 834.684.000001 -5.213 p 5%213.000001-8,806 0 8.806.000001- 65,535 Figure 3.1: River Basin for the study area The areal distribution of the total drainage basin of the Brahmaputra River is given as the following table. 31 Table 3-1: The Brahmaputra River: Country and Indian state-wise S. No. Country Basin Channel Length (km) Arae(km2) 1. Tibet(China) 293000 1625 2. Bhutan 45000 - 3. India A) Arunachal 81424 278 Pradesh B) Assam 70634 640 C) Nagaland 10803 D) Meghalaya 11667 - E) Sikkim 7300 - F) West Bengal 12585 - Total 194413 918 4. Bangladesh 47000 337 The study area includes from Kobo (cross section 65) to Dhubri (cross section 2) which has around 622.73km in length. 32 Figure 3.2: Map of India and study Area of Brahmaputra River 302 Longitudinal Section of Brahmaputra River The longitudinal section of the study reach of Brahmaputra River from cross section 65 (kobo) to Cross section 2 (Dhubri) which is around 622 km is depicted in Figure 3.3. Figure 3.3: Longitudinal Section of the study reach The slope of the river is about 0.27 m/km from Kobo to Dibrugarh, 0.17 m/km from Dibrugarh to Nimatighat (near Bessamora), 0.15 m/km from Nimatighat to Tezpur, 0.14 m/km from Tezpur to Pandu (near Guwahati), 0.11 xn/km from Pandu to Jogighopa 0.094 m/km from Jogighopa to Dhubri and 0.079 m/km from Dhubri to the mouth. 33 40 NUMERICAL MODELING AND FLOW SIMULATION 401 Introduction Rivers are complex and dynamic. It is often said that a river adjusts its roughness, velocity, slope, depth, width, and plan form in response to human activities and (perhaps associated) changing climatic, geologic, and hydrologic regimes. This adjustment may be rapid or slow, depending up on the character of the forces spawning the adjustments. When a river channel is modified locally, that modification may initiate changes in the channel and flow characteristics may propagate both upstream and downstream and through tributary systems. These changes may occur over large distances and persists for long times. Hence the change in river channel capacity leads to flooding. Floods are primary impetuses for all alluvial river morphology. An increase in discharge may initiate bed surface movement and bank erosion, once the force exerted by the flood event (the impetus) has passed some threshold for movement or erosion. This threshold may require a specific flow magnitude and duration before producing a significant morphological response. The timing and frequency of the flood may also have profound effects on a population, a flood may cause damage to civil infrastructure located on or nearby the river. Effective analysis of river problems requires recognition and understanding of the governing process • in the river system. There are two basic items that must always be considered in river hydraulics analysis: the characteristics of the flow in the river, and the geomorphic behavior of the river channel. These two components are sometimes treated separately: however, in alluvial channels (channels with movable boundaries) the flow and the shape of the boundary are interrelated. One dimensional, steady state, fixed—bed water surface profiles are often computed as part of "traditional" river hydraulics. However, some flood plain management, flood control, or navigation studies may require consideration of unsteady (time dependant) flow, mobile boundaries (boundary characteristics that can change with flow and time), or multidimensional flow characteristics (flow with non uniform velocity distributions) to properly perform the required studies ([]SAC Engineering manual, 1993). 34 4.2 Data Sources and Data Types The data used for the simulation of mathematical modeling of the river using HEC RAS software are: 1. Hydrological data 2. Sediment Data 3. Satellite and DEM data 4. Morphological data 402.1 Hydrological data ' Discharge and stage data of Brahmaputra River were collected from two cross sections along the main river channel obtained from Central Water Commission (CWC), Assam Flood Control Department and Brahmaputra Board have constituted main data resources to the model implementation. The available duration of the data was from 2003 to 2007. 4.2.2 Sediment Data Daily sediment data were collected at Jogighopa (Cross Section 9) and Pandu/Guwahati (Cross Section 22) of the river. The data collected from Nov 2003 to Dec 2007 time span are used for simulation purpose. Table 4-1: Types of Data available SL No Cross-section Type of data Duration/Year 1. Jogighopa Daily discharge v/s sediment 27 Nov 2003-31 Dec 2007 2. Pandu Daily discharge v/s sediment 27 Nov2003 — 31 Dec 2007 4.2.3 Morphological data Morphometric data (the reduced levels) of the Brahmaputra river cross-sections of post-monsoon period for the years 1957, 1971, 1977, 1981, 1988, 1993 and 1997 have been collected in respect of all the 64 pre-defined river cross-sections from the Brahmaputra Board, Government of India. 35 4.2.4 Satellite and DEM data Satellite data of the river is available which was collected in the year 1997 and 2008. In addition to this a Digital elevation model data of 90m resolution has been downloaded from http://srtm.cs.i.cgiar.org/ freely. Figure 4.1: Satellite image of Brahmaputra River in 2008 Legend MJoselcOCJtng VALUE ■ 0-446 ■ 446.0000001-1,118 ■ 1,118.000001- 1,833 0 1,033.000001- 2,580 ■ 2,580.000001- 3,357 13,357.000001- 4,075 B 4.075.000001- 4,664 CI 4.664.000001- 6213 O 5,213.000001- 8,806 0 8.806.000001 - 65,535 Figure 4.2: DEM data for Brabonaputra River 36 403 Data Processing The only available data are at the two cross section of the river. But it is required to generate the discharge data at most upstream part of the river. This is because in numerical modeling one of the most important parameter is the upstream boundary conditions. The upstream river location is at cross section 65 with a place called Kobo. At this point there is no gauge site to measure the flow rate. To generate the flow rate at this point drainage area ratio method has been used. The drainage-area ratio method was used extensively in several studies (cliche and others, 1989; Guenthner and others, 1990; Emerson and Dressler, 2002) to estimate monthly stream flow for North Dakota and Minnesota. This method is easy to use, requires little data, does not require any development, and, many times, is the only method available because regional statistics or precipitation-runoff models have not been developed (Douglas G. Emerson, 2005). The drainage-area ratio method is based on the assumption that the stream flow for a site of interest can be estimated by multiplying the ratio of the drainage area for the site of interest and the drainage area for a nearby stream flow gagging station by the stream flow for the nearby stream flow gagging station. Thus, the drainage-area ratio method is given by YU = (?) x11 (4-1) Where, Y is the estimated stream flow, in cubic meter per second, for month i and year j for the site of interest; C' Ay is the drainage area, in square kilometer, for the site of interest; AX is the drainage area, in km2, for the stream flow-gauging station; & X;3 is the stream flow, in cubic meter per second, for month i and year j for the stream flow-gauging station. 37 By using Arc hydro 9 extension tool of Arcgis 9.3 the drainage area of the watershed have been calculated from the delineated DENT of the river watershed. The drainage area at the two points, Kobo and Pandu, has been calculated and the results are listed in the following Table 4.2 below. Lt t1I 2 1 0 2D nnl Degr eg • CS-MGEO) _ - Remet(Gee o) Cetdimcf Ge,COrd Figure 4.3: Catchment area of Brahmaputra River Table 4-2: Drainage Area Ration at different points of the river cross section Cross section Point Drainage Area (km) Drainage Area Ratio with respect to Pandu Dhubri(CS 2) 423469.58 Pandu(CS 22) 324309.94 Bessamora(CS 50) 202712.23 0.62 Kobo(CS 64) 184238.43 0.56 44 Model Development For this study quasi unsteady flow simulation has been used for the development of the model for Brahmaputra River. Yang, Engelend-Hansen and Ackers-White method of transportation methods are used for the development of the model and calibration of sediment. 38 4.4.1 Data Requirements and input The basic data requirement for modeling of the river using HEC RAS are 1. Hydrological data 2. Sediment data 3. Observed data for calibration 4. Morphological data 4411 Hydrological Data The hydrological data required for the analysis of the model depends on the type of flow used either steady flow, unsteady flow or quasi unsteady flow. Steady flow is used for calibration of the roughness coefficient and Quasi unsteady flow is used for the general development of the river model Quasi-unsteady flow has been used for sediment transport analysis. Flow series, rating curve and lateral flow series are prepared for the simulation of the model in quasi unsteady flow condition, which are inputted as boundary conditions. Flow series at the upstream part of the river, i.e. at cross section 65 (Kobo) is used as upstream boundary condition and at the downstream part of the river rating curve have been prepared. In addition to upstream and downstream boundary conditions internal boundary conditions are used at Pandu and Jogighopa. Flow rating curve at Dhubri is shown in figure 4.7 below. Flow series at kobo is based on the generated data from Pandu from 27 Nov, 2003 to 31 Dec, 2007. 39 35000 30000 Flow Series @ kobo 25000 20000 m 00 15000 N 0 10000 5000 0 27-Nov-03 14-Jun-04 31-Dec-04 19-Jul-05 4-Feb-06 23-Aug-06 11-Mar-07 27-Sep-07 Time (Days) Figure 4.4: Flow series at Kobo 25000.00 Lateral Flow @ Pandu 20000.00 m 15000.00 10000.00 5000.00 0.00 I 27-Nov-03 14-Jun-04 31-Dec-04 19-Jul-05 4-Feb-06 23-Aug-06 11-Mar-07 27-Sep-07 Time (days) Figure 4.5: Lateral Flow Series at Pandu 25000.00 TT 1!Th.ii 20000.00 15000.00 0 10000.00 Imiril 5000.00 __ ['A ______ 0.00 27-Nov-03 14-Jun-04 31-Dec-04 19-Jul-05 4-Feb-06 23-Aug-06 11-Mar-07 27-Sep-07 Time (days) Figure 4.6: Lateral Flow Series at Panda 40 For the development of the rating curve the monthly discharge collected from 1996 to 2002 at Dhubri site is used. The method used for the development of rating curve is Q = cc (G — Gm)D (4.2) Where: Q is flow rate in m3/s G is the stage in meter Go is the gauge at which the flow is 0 and a and (3 are constants which are computed by taking logarithm of the above equation from different discharge and stage relation. The value of Go is calculated by using Jonson Method (WMO Operational Hydrology manual on stream gauging (Report No. 13, 1980)). The other coefficients a and 13 are determined by least square method. 41 Table 4-3: Discharge and Stage Relation at Dhubri Discharge, Stage, Q(m3/s) S(m) 000 1994 147.06 21.00 38800 4°° 215' 779.76 22.00 1345,48 " ' 22.5O 35.00 2105.65 23.00 ~~-3078.87 {' _;" 23.50 , 30.00 4282.27 24.00 25.00 5731.83: 245Ø ' 20.00 7442.61 25.00 tn2115.00 .9~428.86 x 25.50 10.00 11704.16 26.00 5.00 1428151 26.50 > 0.00 17173.38 27.00 0.00 15000.00 30000.00 45000.00 60000.00 Discharge (m3/s) 0391.79 ; 27.50 23948.34 28.00 Figure 4.7: Rating Curve at Dhubri =27854`27 28.50 '. 32120.49 29.00 r 29.50 41775.85 30.00 4;47185.36." 30.50 52995.91 31.00 X5921707 31.50 65858.22 32.00 42 4.4.1.2 Sediment Data Sediment data are useful data to simulate the system by mobile bed condition in HECRAS 4.1. Sediment data includes inflowing sediment load data, gradation of material in the stream bed and information about sediment properties. The inflowing sediment load is related to water discharge by a relating table at the upstream end of the model. An alluvial channel system responds continuously to changes of discharge, grain distribution, or other parameters. It does so by modifying the flow field and the channel geometry mainly by changing the bed form. The system tends to create a state of equilibrium, which is another way of saying that nature always tends to minimize an existing gradient. A change in the bed form must be due to a transport process. At one place sediment is eroded to be transported and deposited at another location in the channel. The transport needs the fluid as carrier, and the interaction is governed by a feedback process. Each cross section must have an associated bed gradation. The percentage of fines is used for the formulation of the model which is collected from the site. Bed gradation sample was taken from "Palasbari" site downstream of Guwahati (Pandu). Diameter of sediment particles and percentage finer are tabulated and plotted as follows in Table 4.4 and Fig 4.8. 43 Table 4-4. Sediment gradation at Palasbari Diam Cumulative S no Class (mm) %Finer %Finer Fraction 1 Medium Silt(MM) 0.032 0 0 0 2 Coarse Silt(CM) 0.0625 8 8 0.08 3 Very Fine Sand (VFS) 0.125 10 2 0.02 4 Fine Sand (FS) 0.25 36 26 0.26 5 Medium Sand (MS) 0.5 55 19 0.19 6 Course Sand (CS) 1 82 27 0.27 7 Very Course Sand (VFS) 2 87 5 0.05 8 Very Fine Gravel (VFG) 4 90 3 0.03 9 Fine Gravel (FG) 8 93 3 0.03 10 Medium Gravel (MG) 16 96 3 0.03 11 Coarse Gravel (CG) 32 98 2 0.02 12 Very Course Gravel (VFS) 64 100 2 0.02 !fIA!1l11!® , -®lhHI Il11lul!!®aI II, !lIll1V.!!!1l11IRI!!I .- ., !!!1w/!!1lin®!!i lP1-!!1l18U!!I , !!.fl1!!1l1I!!!R ®1111!!1l111!!!! , _!!1 UIl111!!!!! „ , Figure 4.8: Representative Bed Gradation (Semi log) plot of the study reach 44 Sediment boundary used includes sediment rating curve, sediment load series and equilibrium load. At the upstream boundary condition equilibrium load is used. Sediment rating curve are developed from the observed data for internal boundary conditions at Pandu and Jogighopa. Discharge and sediment data are available at the two cross sections. The discharge vs sediment concentration relationship at the two cross sections has been approximated by using different equations. At. cross section 22 a third order polynomial equation "y = 7*10-8x3 - 0.001x2 + 115.3x — 39977 is used. Similarly at cross section 9 (Jogighopa) the discharge and the sediment concentration data are related by using third power equation of y = -5*"8x3 + 0.004x2 + 4.783x + 25740. Where y is sediment concentration in tons/day and x is the discharge in m3/s. 1.80E+07 1.60E+07 1.40E+07 1.20E+07 c 0 N 1.00E+07 8.00E+06 o :. 6.00E+06 4.00E+06 2.00E+06 0.00E+00 0 10000 20000 30000 40000 50000 60000 Discharge (m3/s) Figure 4.9: Sediment Rating Curve at cross section 22 (Pandu) 404.103 Morphological and Geometrical Data Geometry of the physical system is represented by cross sections, specified by coordinate points- (stations and elevations), and the distance between cross sections called downstream reach length. The basic geometric data consists of establishing the connectivity of the river system (river System Schematic); cross. section data; reach lengths; energy loss, coefficients (friction losses, contraction and expansion losses); and stream junction information. Hydraulic structures data like bridges, culverts, spillways, weirs, etc are also grouped as geometric data. 45 The hydraulic roughness is measured by the manning's coefficient. The value of the manning coefficient may be different at different cross sections and also varies with time. Spatial and temporal variation of the coefficient has to be analyzed thoroughly before -using it in the model. The river system schematic is required for any geometric data set within NEC RAS. It is developed by drawing and connecting various reaches within the Geometric editor. If there is DEM data of the river, it is possible to import geometrical data of the river from Arcmap. This data is developed by using HEC-GeoRAS extension in ArcGis. The river center line and the bank lines of the river have been imported using HEC GeoRas tool to DEC RAS. Figure 4.10: Geometric Profile of Brahmaputra River with interpolated cross section Each cross section in an NEC RAS data set is identified by a river reach, and river station level. The cross section is described by entering the station and elevation from left to right with respect to looking in the downstream direction of the river flow. The river stations identifier may correspond to stationing along the channel, mile points, or any fractious numbering system. The numbering system must be consistent, in that the program assumes that the higher numbers are upstream and lower numbers are downstream. 46 4.402 Energy Loss Coefficient Several types of loss coefficients are utilized by the program to evaluate energy losses like manning's n values or equivalent roughness "k" values for friction loss, contraction and expansion coefficients to evaluate transition (shock) losses, and bridge and culvert loss coefficients to evaluate losses related weir shape. Manning's n: selection of an appropriate value for manning's n is very significant to the accuracy of the computed water surface profiles. The value of manning's n is highly variable and depends on a number of factors including, surface roughness, vegetation, channel irregularities, channel alignment, scour and deposition, obstruction, size and shape of the channel, stage and discharge, seasonal changes, temperature, and suspended material and bed load. With the initial value of Mannings roughness coefficient the model is run for the first time and once the stability of the model is seen the next step is to calibrate the roughness coefficient. Manning's n values are calibrated in fixed bed condition based on discharge variation and values fed into sediment module of flow simulation for further analysis. 4.5 Results and Discussions 431 Calibration of Bed Roughness Calibration is the adjustment of model's parameters, such as roughness and hydraulic structure coefficients, so that it produces observed data to an acceptable accuracy. Roughness coefficients are one of the main variables used in calibrating a hydraulic model. Observed flow series of year 2004 at Jogighopa (CS 9) and Pandu (CS 22) are used to calibrate the roughness coefficient n. The calibration is done for the range of flows for each month and hence roughness variation for high and low flows are accounted. Different roughness coefficients have been taken and verified with the observed value of at the two sites of the river. Monthly flows are taken and rigid bed condition is used for calibration purpose. 47 At cross section 9 (Jogighopa) Manning's n value of 0.04 and at cross section 22 (Pandu) Manning's n value of 0.03 have good fit between observed and computed water surface elevation as shown in the following figures and tables. 39 37 35 5 33 I 31 29 27 J F M A M J J A S O N Time (month) Figure 4.11: Observed and Simulated values at cross section 9 (Jogighopa) 48 146 co 44 3 42 e W40 J F MA Ml 1 J A SON Time Figure 4.12: Observed and Simulated Values at cross section 22 (Pandu) From the above results the correlation coefficient and standard error of the simulated and observed values at the tow gauging stations for the above values of Manning's n are tabulated as follows: Table 4-5: Goodness of fit observed and simulated water surface elevations Station Correlation Coefficient(R) Standard Error Jogighopa 0.9912 0.3663 Pandu 0.9966 0.1824 48 For the development of model for Brahmaputra river a Manning's n value of 0.04 and 0.03 and 0.035 are used for different cross section based on the calibration. 4.5.2 Sediment Transport Analysis Using the calibrated manning's roughness value mobile bed sediment transport has been done for four year simulation periods from 2004 to 2007. The analysis was done using three different transport functions, Ackers-White, Engelend-Hansen and Yang methods. Ackers-White is developed in flume for overall particle diameter of 0.04-7 nun and young developed his equation based on filed observation of overall particle diameter of 0.15-1.7 mm, which bases transport on stream power, the product of velocity and shear stress. Engelend-Hansen is developed based on flume data for particle mean diameter of 0.19 mm to 0.93 mm. It has been extensively tested and found to be fairly consistent with field data (HEC-RAS Manual, 2010). Transport potential was calculated at each cross section using hydraulic information yielded from the water surface profile calculation and the gradation of bed material. Sediment was routed downstream after the backwater computations were made for each successive discharge (time step). The theoretical basis for adjusting bed elevations for scour or deposition was provided by the continuity equation for sediment material or the Exner equation. 1. Water Surface Elevation The simulated and observed water surface elevations at Pandu (Cross Section22) and Jogighopa (Cross Section 9) gauging sites are used as verification of the model. The graphs of water surface elevation of the observed and computed values are plotted in Appendix H using different method of sediment transport functions. The correlation coefficients between these values are given in the following Table 4-6. Table 4-6: Goodness of fit for water surface elevation Ackers-White Yang Method Engelend-Hansen Gauging site R S error R S error R S error Pandu (CS 22) 0.967 0.561 0.972 0.508 0.972 0.509 Jogighopa (CS 9) 0.989 0.317 0.962 0.595 0.983 0.403 49 2. Sediment Transport Capacity Ackers-White method of transportation yields higher values of sediment transport capacity during high flow and lower values of sediment transport capacity during low flows compared to the other transportation functions. On the other hand during high flow seasons Yang method of transportation gives better fit to the observed values. During low flow season the variation becomes high with the computed and observed values. The goodness of fit on Sediment Mass capacity/Sediment Volume capacity between computed and observed flow, at the two gauge site of the river for different transport functions are shown in the following Table 4-7. Table 4-7: Sediment Capacity Goodness of fit at Pandu and Jogighopa Ackers-White Yang Engelend-Hansen Gauging site R R R Pandu (CS 22) 0.551196 0.900848. 0.899164 Jogighopa (CS 9) 0.382025 0.689397 0.701768 3. Bed Change AckersWaite Method of Transportation Function Due to sediment transport there is a change of bed level of the river through time. The plot in Fig 4.13 shows that there is change in bed level across the cross section starting from upstream of the cross section to the downstream section of the river. Maximum aggradation is around 6.4m at cross section 25 and maximum degradation of 1.87 m is observed at cross section 22. The invert changes at different cross section of the river are shown in the following Table 4-8. 50 Table 4-8: Change in invert Elevation at different time Invert Change (m) RS 22 Jul, 2004 20 Aug, 2007 29 Dec, 2007 Remark 54 -0.250 -0.490 -0.5043 Degradation 52 -0.052 -0.127 -0.1356 Degradation 51 -0.068 -0.156 -0.1636 Degradation 50 0.862 1.781 1.8325 Aggradation 46 1.968 2.617 2.5407 Aggradation 45 0.175 0.495 1.0088 Aggradation 43 0.042 1.277 1.4145 Aggradation 42 0.027 0.506 0.5000 Aggradation 41 -0.017 -1.165 -1.1679 Degradation 36 0.008 3.090 3.5162 Aggradation 34 0.026 0.616 0.5444 Aggradation 32 0.038 1.615 1.6072 Aggradation 31 0.059 1.541 1.2921 Aggradation 30 0.123 1.792 1.8184 Aggradation 29 0.051 3.136 3.6569 Aggradation 28 -0.003 0.124 0.4415 Aggradation 27 0.001 3.094 3.2501 Aggradation 26 0.000 0.518 0.7156 Aggradation 25 0.001 .5.298 6.4152 Aggradation 23 -0.024 -0.058 -0.0599 Degradation 22 -0.019 -1.837 -1.8734 Degradation 20 -0.001 4.781 5.1855 Aggradation 18 0.000 1.349 1.4846 Aggradation 17 -0.005 -0.016 -0.0176 Degradation 9 -0.013 -1.229 -0.6728 Degradation 6 0.000 0.080 -0.0138 Degradation 5 -0.021 -0.054 -0.0576 Degradation 4 0.004 1.216 1.8001 Aggradation 3 0.000 0.060 0.0925 Aggradation 2 0.002 1.370 1.5661 Aggradation Max 1.968 5.298 6.4152 Aggradation Max -0.250 -1.837 -1.8734 Degradation 51 d:1DE MlBdMAS Rawl% and simulation*24 hrlime syanWytsm(A4h.a3d01 &et..~aan.an n-sn _ J ne ' o: J1w~Wm0oLDYwlLlagel t i y fE ~DBC1'UOIOammh•ed Cnao.UN t i [ t •~ 111 4 €tt o +oo 20o mo ®a mo ma 700 WInC1an6Dirbioo Figure 4.13: Change in Invert (m) along the river starting from Cross section 2 (0.0) Yang's Method This method of transportation yields a maximum degradation of 1.5 m at cross section 54 and maximum aggradation of 0.851 m at cross section 50 during four year simulation period in Brahmaputra river. Graphically the degradation and aggradations across the length of the reach are shown in the following Fig 4.14. MADE MUC:Vih9 R.d:band almutaCOnd24 hrdm.,.n1y.SA,(YA)md03 mtmm.~.~nsn i mswm+mmnwMa.v.1.V _ ` 9JUl~sOUmat•1MLTn0.1:N ...... mm mm ~ • :d 1 g n 7 -~ o m an mo 0o mo ma ma Figure 4.14: Change in Invert (in) along the river starting from Cross section 2 (0.0) using Yang Table 4-9: Degradation at some cross section at different times Invert Change (m) RS 22JUL2004 04AUG2005 04FEB2007 29DEC2007 Remark 54 -0.198 -0.890 -1.313 -1.485 Degradation 53.3333* 0.043 0.221 0.319 0.356 Aggradation 53 0.015 0.097 0.180 0.226 Aggradation 51.5* -0.054 -0.215 -0.463 -0.565 Degradation 52 51 -0.023 -0.257 -0.554 -0.652 Degradation 50.3333* 0.042 0.277 0.667 0.851 Aggradation 50 0.015 0.185 0.543 0.727 Aggradation 47 -0.010 -0.038 -0.074 -0.092 Degradation 46 0.005 0.021 0.046 0.065 Aggradation 41 -0.020 -0.072 -0.162 -0.273 Degradation 39 -0.016 -0.047 -0.086 -0.117 Degradation 38 0.041 0.108 0.217 0.302 Aggradation, 37.25* -0.009 -0.030 -0.055 -0.067 Degradation 37 -0.006 -0.014 -0.033 -0.039 Degradation 33.25* -0.011 -0.034 . -0.069 -0.078 Degradation 33 0.000 -0.008 -0.017 -0.020 Degradation 22 -0.047 -0.166 -0.345 -0.508 Degradation 8.33333* -0.006 -0.019 -0.053 -0.067 Degradation 7.5* 0.006 0.016 0.030 0.035 Aggradation 4.33333* 0.006 0.020 0.039 0.052 Aggradation Max Aggradation 0.043 0.277 0.667 0.851 Max Degradation -0.198 -0.890 -1.313 -1.485 Engelend-Hansen Method In this method the maximum decrease invert elevation simulated during the four year simulation period is 1.3 m at cross section 54 and a maximum increase of invert elevation 0.85 at cross section 50. Graphically the change in channel invert starting from cross section 2 to the upstream cross section is shown in the following plot on Fig 4.15. 53 d:IDE MleohlRAS Rea~itsand sm,1aion924 hrtlme ;an14yr51m(EH}md03 Brtlm~pbsll an Re¢h mA . 3 _ _ .. _M1...,. ~..- 'ILa111fO1O21RtM111bILls6e(m) - . I,._£. ~040¢®-halCMOOlm1.. 0.0 210 90 4 (flp ®D RO Figure 4.15: Change in Invert (an) along the river starting from Cross section 2 (0.0) From the above results it "can be conclude that Ackers-White method of transportation gives a high value of sediment transport capacity hence the change in bed elevation through the simulation period reaches up to 6.4 m of aggradations. On the other hand Yang's method of transportation has good fitness values of sediment transport capacity with the observed values at Pandu and Jogighopa stations having a correlation coefficient of 0.9 and 0.7 at the two gauging sites Pandu and Jogighpa. 54 50 SIMULATION STUDIES FOR CHANNELIZATION OF BRAHMAPUTRA RIVER 501 Introduction Rivers have been a focus of human activity throughout ancient and modern times. So important to humanity are the benefits obtained from rivers, and so necessary is the protection against floods and other river disasters, that pursuit for knowledge of riverine systems has advanced in leaps and bounds. While engineers are interested in water supply, channel design, flood control, river regulation, navigation improvement, and so on, it is clear that rivers, as a part of nature, can be mastered not by force but by understanding. Rivers have been a subject of study by engineers and scientists who have been fascinated by their self formed geometric shapes and their responses to changes in nature and human interference. In addition to engineering, understanding river behavior is also necessary for environmental enhancement. Generally alluvial rivers which cause damage due to high flood can be controlled and it is possible to minimize the risk of causing loss of life and loss of the properties and infrastructures of human beings. In addition to that river banks need to protected from erosion due to flood. This thesis includes analyzing the effect of channelization works at specific nodal points. At this time flooding is the major problems in the area of flood prone areas. So channelization of a river has a major impact on the minimization of flood problem and reclamation of land which are inundated during floods before channelization. The use of river channelization is not only for flood protection but also for different purposes as listed in section. 502 Types and Methods of Channelization Channelization of rivers and bank protection works are taken up to achieve several objectives which are listed as follows; Flood Control and Protection.- Undulation of water beyond the bank of the river cause several problems, like loss of life, erosion of the earth surface, damaging of infrastructures, loss of individual and public properties, loss of agricultural lands. Construction of Leaves and embankments, increasing the discharge capacity of rivers by widening or deepening of the river 55 channel, constructing of escape channels from the main river to decrease the flow in the main river are river training structures used to control the flood. Guiding the River Flow: Some alluvial rivers are highly braided in which the width of the river is very wide and difficult to manage. To construct a bridge across these braided rivers will be very costly and sometimes impossible due to the nature of the river. River training structures along the river to guide the flow and to protect the banks of the river are constructed. Sediment control: Sediment transportation by the river from the banks of the river and from the bed of the river is the major problem. To protect the river from sedimentation river training structures like spurs at the bank of the river are provided as a river training structures. Stabilization of River Courses: The river banks are weak and can be easily eroded and caved due to the flood passing through it. To protect this river training structures play a crucial role in protecting of them. There are different types of river training works used for different purposes such as leaves, spurs/groyens, and guide banks etc. depending on the type of purpose used, different channelization are used. Resectioning by Widening and Deepening: -. > Straightening:- Levees (or Embankments):- > Flood Walls and Lined Channels: - Revetment/ Bank Protection Structures: - > Diversion Channels: - > Culverts: - > Groynes/Spurs: Guide bunds/Gide Banks: Different methods of channelization can be used at different cross section. In case of Brahmaputra river the following methods of channelization can be adopted. 56 1. Guide bunds, mostly used for confine the flow especially used when there are hydraulic structures like bridge, and others. 2. Closure of secondary channels by triggering sedimentation of river training structures like porcupine screens, Jack Jetty, permeable spur etc. 3. Using Geotextile tubes for control of bank erosion, and partial closing of channels in phases. Design Consideration of flow guide bunds The waterway is the actual width through which flow takes place at the structure. Generally, the length of clear waterway provided between the guide banks is taken equal to Lacey's regime paiameter. P=4.75/ (5.1) Where: P is Lacey's water way perimeter (m) and Q is the maximum discharge (m3/s). In case of Brahmaputra River the water way becomes too small when it is compared with the braded pattern of the flow. With especial consideration of morphology of the cross section higher values of waterway than Lacey's have been used. Top Level of the guide banks is kept equal to the upstream total energy level plus adequate free board. 2 Top Level of the Bank = HFL before consrtuction + aflux + Zg + Free Board (5.2) Where: afflux is the rise in water level n the upstream of the structure. Afflux is maximum near the structure and gradually decreases as the distance from structure increases in the upstream direction. A free board of 1.25 m to 1.50 m is above the maximum anticipated upstream high flood level is usually provided for guide bunds. Length of Guide banks: according to spring, the length of the guide bank on the upstream side measured from the axis of the structure should be equal to 1.10 L and that on the downstream side between the 0.10 L to 0.20 L. Where: L is the overall length of the water way between the 57 abutments. According to Gales, the lengths of the upstream guide should be taken according to the river discharge (Arora, 2002). Table 5-1: U/s and d/s length of the guide banks Gale's guide line Discharge (m /s) Length of u/s Guide Convergence towards bank structure Less than 21000 1.25 L 1 in 20 21000 to 42500 1.25 L to 1.50 L 1 in 20 to 1 in 40 Greater than 42500 1.50 L 1 in 40 The length f the d/s guide bund is recommended as 0.25 L for all discharges (Arora, 2002). 5.3 Hydrological and Sediment Data Flow parameters and sediment transport capacities are compared before and after channelization works at selected cross sections of the river. From the observed data of years 2004 to 2007, maximum flow was recorded during 2004. Maximum discharge at Pandu during this year is 55235.21m3/s. This year flow data has been used for simulation of parameters using HEC RAS 4.1.0 to compare the results before and after the construction of training structures. Flow series, lateral flow series, and rating curve are used as flow boundary conditions at upstream cross section, internal cross sections and downstream cross section respectively for quasi unsteady flow. The sediment rating data is used as sediment boundary conditions at the two internal boundary conditions (cross section 9 and 22). Equilibrium load is used as upstream boundary condition at cross section 65. 58 Sed RC @pandu 1.60E+07 1.40E+07 V 1.20E+07 1.00E+07 C 8.00E+06 6.00E+06 E 4.00E+06 2.00E+06 0.00E+00 0.00 10000.00 20000.00 30000.00 40000.00 50000.00 60000.00 Discharge (m3/s) Figure 5.1: Sediment Rating Curve at Cross section 22 (Pandu) Rating Curve@cs9 5.00E+06 4.50E+06 4.00E+06 3.50E+06 3.00E+06 2.50E+06 0 2.00E+06 U 1.50E+06 1.00E+06 5.00E+0S 0.00E+00 0.000 15000.000 30000.000 45000.000 60000.000 75000.000 Discharge (m3/s) Figure 5.2: Sediment Rating Curve at cross section 9 (Jogighopa) Geometrical data of the channel at each case will be different based on the alignment of the guide banks and other river training structures at each cross section taken. The guide bunds may be constructed on both banks of the river or on one side of the river depending up on topography of the river banks. 59 504 Channelization at Different Nodal Points 5.4.1 Introduction Brahmaputra River is a highly braided river in which some channelization works has to be done to confine the width of the flow for construction of bridges, navigation and to reclaim the land which are part of flood plain. Along the river 64 cross sections are surveyed at different times of the year. Recently surveyed cross sections are used for simulation purpose of the model. From the available data it is observed that the maximum width of the river is at cross section 29 which is around 22 km. Similarly the minimum width of the river is at Pandu (cross section 22) which is reaches around 1.5 km. 25 20 E 15 to 5 0 2 6 10 14 18 22 26 30 34 38 42 46 50 54 58 62 Cross section Figure 5.3: Surveyed width at different cross section of Brahmaputra River It will be difficult to put channelization/river training structures over the whole river from economical point of view. Some nodal points are selected along the river for channelization of the river. After providing channelization structures along the banks of the river like guide bunds or other river training structures, their model was simulated and the result before the construction of the channelization work has been compared. At some cross section of the river bridges are required to connect places in the opposite banks of the river. Constructing of a bridge across the full width of the river may not be economical because of large width of the river. Confinement of the river flow is mostly provided to avoid such problems by constructing river training structures. 504.2 Channelization at Cross Section 57 Cross section 57 is near Dibrugarh and geographically located at 27°24'46.98"°N and 94°45'19.59"E. There is a bridge at this place which is under construction for railway transport. The length of the bridge is around 4 km and the width of the river at this cross section is around 9.5 km. A guide bund is constructed on both sides of the bank to confine the braided flow patern of the river to 4 km and to guide the flow of water smoothly to the bridge openings. The dimension of the guide bunds depending on Gale's guide line from Table 5-1 above is shown below. Length of the Waterway =4 km U/s Length of the guide banks = 1.5 *4=6 km D/s length of the guide bank = 0.25*4 =1 km Convergence towards the structure with I in 40 Total length of the guide bank =7 km on both sides of the bank. Using .HEC-RAS 4.1.0 the change in water surface elevation, velocity, sediment capacity and other parameters have been checked before and after channelization. The effects of each parameter are discussed in the following sections. Figure 5.4: Satellite images Cross section 57 61 5..4.201 Effect of Channelization at CS 57 Water surface elevation: There is an increase of water surface elevation upstream of cross section 57 at which the bridge center line is located. This change is due to the afflux of the flow. The rise of the water surface extends up to cross section 60 which is around 30 km from cross section 57. Graphically the simulated values of the water surface elevation at each cross section of the river before and after guide banks at high flow are shown in Fig 5.5. The maximum change in water surface elevation during high flow period is 0.81 mat cross section 57.6666* which is 7.481 km from cross section 57. —$"-WS.EIevBGB —i-WS.EIevAGB 0 110 108 106 4- 104 102 56 57 58 59 60 61 62 Cross section Figure 5.5: Water surface elevation before and after Guide bund at July 22, 2004 62 Table 5-2: Water Surface Elevation and Velocity before and after Channelization upstream and downstream of cross section 57 River W S Elev (ffi) Change Velocity (mis) Change is station in Elev Velocity (m) 60.75* 110.05 110.05 0 0.97 0.97 605 10905 090' 002 0962 0.95 .Sq{ F~ n 001 ~1 60.25* 108.29 108.34 0.05 0.87 • 0.85 -0.02 y ~ 4 60 10785 (107.92 ~y`007 0.73 0.71 k '`v ®02 ~ ~• 59.6666* 107.47 107.58 0.11 0.79 0.76 -0.03 87 T ~.'...~,.~;~''i~'.~''z~.;' 59 106.44 106.7 0.26 0.98 0.91 -0.07 ~'~ .f'~z"' `'.~ 6 * 105.84 0 43 0.95 ~ 58 1 0.8~ ~,... a ._.. 58.3333* 105.37 105.97 0.6 0.88 0.75 -0.13 58 :'105.04 105,77 «yx~~ 0,73 0.78 . 0:66 •012 57.8333* 104.87 105.67 0.8 0.8 0.67 -0.13 m-„.r t ~+, ~> ,~ ~' 3 .yam '~ 'i "s y ~nra ~: ~.~ 7 x 5`16666* 10468 fi 10.5 49 ° 08i 083 ✓ 098 0J5 5705* 104.49 105.21 0.72 0.86 1.23 0.37 57.1666* 104.06 104.48 0.42. 0.95 1.38 0.43 57 103.8110. 0.21 102 1 ,5 r x 048 56°9444* 103.66 103.74 0.08 1.02 1.54 0.52 6888 103.52 x 103.53 i 0.01 1.02 25 0.2 ....✓ ... 56.8333* 103.37 103.37 0 1.03 1.03 0 63 Velocity Change. The velocity of flow decreases in the upstream section of the river station up to cross section 60.25* as shown in the above table and it starts to increase from cross section 57.6666* up to downstream section of 56.8888* during maximum flow. The maximum velocity at this time is 1.5mis which occurred after channelization at cross section 57. Then it starts to decrease downstream. Graphically the change in velocity across the cross section is shown in the following graph on Fig 5.6. —4--VelocityBGB ---VelocityAGB 1.8 1.6 1.4 7 1.2 ~ 1 j 0.6 0.4 0.2 0 56 57 58 59 60 61 62 Cross section Figure 5.6: Cross section Vs Velocity before and after construction of Guide bank At cross section 57 the waterway width is decreased from 9.5 to 4 km. Due to decreasing the width of the cross section by constructing guide bund, velocity time series increased. There is an increase of velocity from 3% to 32.44% during the simulation period. The variation in velocity. before and after channelization of the river at cross section 57 is shown in the following graph on Fig 5.7. 64 Before Channelization After Channelization dDEp.etRpS WDEMtwhIRAS Rwowlta d SknAd- Sow ti Lgmd lb 12 1.01 .?an Feb Mar Apr May .Ami .i ALO Sep O1 Now Dec Jan FebMar Apr May .km Ad ALU Sep Oct Now Dec WM TVM Pum Figure 5.7: Change in velocity through the simulation period at cross section 57 Shear stress There is an increase of shear stress at this cross section during high flow season up to 49.8% after channelization. The shear stress plots before and after channelization is as follows in the following Fig 5.8. Before Channelization After Channelization Figure 5.8: Shear stress time series at cross section 57 Mass Bed Change and Mass Capacity. Due to channelization at this cross section, there is an increase of mass bed change. Mass bed change is the difference in sediment mass in and mass out. It is observed from the simulated value that the increase in mass bed change reaches around 83%. The mass bed change at this 65 cross section implies that there is bed erosion at this point throughout the simulation period as shown in the following plots but erosion is higher in case of channelized condition. Before After d\DE M t cMRAS Rh. and sinid2tiom%RvSimWatIon1WlO AG&snm3 Figure 5.9: Mass bed Change at CS 57 before and after channelization There is also bed change due to the construction of guide bund at this cross section. The change in one year simulation time period indicates that the bed changes are small. This is due to the deep cross section and the distance of the water way is around 4 km and it is simulated for one year flow. The bed material and geological conditions have also an impact on the mass bed change. The mass capacity in tons/day has been increased by 83% due to channelization. This mass capacity can be also converted in to volume capacity in, m3/day. Before After to Figure 5.10: Mass Capacity at cross section 57 before and after channelization 66 Sediment output can be also expressed in terms of volume. Converting between mass and volume is multiplying or dividing by the unit weight of the material. 2500 2000 en E 1500 WEa 1000 E 2 500 0 0+= 1-Jan 20-Feb 10-Apr 30-May 19-Jul 7-Sep 27-Oct 16-Dec Time Figure 5.11: Sediment Volume Capacity before channelization at CS 57 14000 12000 10000 8000 U f0 6000 E 4000 2000 1-Jan 20-Feb 10-Apr 30-May 19-Jul 7-Sep 27-Oct 16-Dec Time Figure 5.12: Sediment Volume Capacity at cross section 57 after guide banks 5.422 Downstream and Upstream Effects Due to channelization of the river effects have been observed in downstream and upstream cross sections from cross section 57. 67 Time series out puts of water surface elevation indicates that there is an increase of water surface elevation in the upstream cross section of the river from cross section 57 due to channelization at this point. Increase of water surface extends up to cross section 60 as shown in the Fig 5.7 above. Velocity of flow shows changes up to cross section 56.833* which is up to the distance 3.0 km from cross section 57 in the downstream cross section. In case of upstream cross sections the velocity starts decreasing from cross section 57.8333* up to 60 during high season as shown in the Fig 5.8 above. The velocity time series for both conditions are listed in Appendix A. Similarly there is an increase of shear stress in the downstream cross section up to 41an. after cross section 56.7777* the values of shear stress are the same before and after channelization. The bed gets eroded from cross section 57 and continues erosion of bed up to 56.8888* i.e 2.0 km away from cross section 57. From cross section 56.8888* to 56.7777* for around 2.0 km„ deposition of sediments starts. Then mass bed change and mass capacity have the same values with the original values before channelization. The plots of mass capacity and shear stress before and after channelization in the downstream part of the cross section are shown in Appendix A. Before Channelization After Channelization d:WE Mt.d RA$ R..'It...d dm.l.eoneltod SI-1 d. DE MI.d.'RAS N.s+It—d dmul.tlondRov Slmol.tlonlWlfhG8~57R.v.md03 SnlJm 1.bs Btla~p.M(Yn) ~ a , .An .L AM S Od Mo, Dc M Mby 1: F. U. M Ikn ,.., T- Figure 5.13: Mass bed change before and after channelization at cross section 56.7777* Mass capacity in tons per day at different cross section in the downstream and upstream cross sections is shown in Appendix A. The results shows that there is an increase of sediment mass capacity in downstream cross section up to 2 km distance. 68 50403 Channelization at Cross Section 55 Cross section 55 is located at a distance of 27.03 km downstream of cross section 57. and geographically located around at 27°13°19.26" N latitude and 94°35'16.63" E longitude. Figure 5.14: Cross Section 55 At this cross section it is proposed to confine the river flow to 2.0 Km from 13.6 km by providing flow guide bunds on both sides of the bank. The dimension of the guide banks depending on Galas Guide lines are calculated as follows. Length of the waterway = 2 km U/s Length of the guide banks = 1.5 * 2=3 km D/s length of the guide bank = 0.25*2 = 0.5 km Convergence towards the structure with 1 in 40 Total length of the guide bank = 3.5 kin on both sides of the bank. 5.4,3.1 Effects of Channelization at CS 55 Water Surface Elevation: Water surface elevation due to the construction of guide bunds to confine the flow at this cross section has increased in the upstream cross sections during high flow season. The maximum change in water surface computed from the simulation result is 1.8 m at cross section 55.4* which is 4.6 km distance from cross section 55 at July 22, 2004. Water surface elevation before and after guide banks are elaborated in the following graph. ---WS EI AGB —*—WS EI BGB 104 — 103 E 102 a1 101 100 v's 99 I 98 39 97 96 54.5 55 55.5 56 56.5 57 Cross Section Figure 5.15.- Water surface Elevation change upstream of cross section 55 Velocity change There is increase of velocity due to placement of guide bund on both sides of the river banks at cross section 55 during high flows. The velocity at this cross section increases from 0.85 rn/s to 2.62 m/s. The change in velocity in downstream and upstream of cross section 55 is shown in the following Fig 5.16. —+—V before Channelization —s—V after Channelization 3 — 2.5 54.5 55 55.5 56 56.5 57 Cross section Figure 5.16: Velocity (Change at upstream and downstream from cross section 55 70 Through one year time series simulation period, there is an increase of velocity after channelization works. It reaches up to 66% increase during high flow season. The plot of velocity time series before and after channelization is show in the following plots. Before After d EE Mt.di1RAS R.alb.nd tlmw.tIonOR., d.1t UINIRAS R.aJb.n!fimulatlenR.v awwm — — b — .fin NG to Oa. to to Figure 5.17: Velocity time series at cross section 55 Shear Stress The sheer stress at this cross section increased in high flow and decreased in low flow season. This is due to morphology of the cross section which is selected for the construction of the guide bank. There is up to 84.5% increase of shear stress due to the construction of guide bank in both sides of the deep area at cross section 55 in high flow season. This particular figure is occurred at 22 July 2004 at which the maximum flow is measured. During low flow season the shear stress decreases up to 21.55% at Dec 5, 2004. This is due to smooth flow of water guided by flow guide banks. The variation of the shear stress can be seen in the following plots. Before After QIDE Mlsh1RAS R.Wb wE dmlrtom%RwSYnJ.Ibr*bMtGB®55Rwu.d LI^ SnWa9m 4 Sh- Mm AW May Jim SP Oct Na Ors i figure .ii: sHear stress at cross section 55 71 Bed Change and sediment Capacity Sediment transported at this point increases due to the construction of the guide banks during flooding period. The maximum increase in mass capacity during high flow times reached up to 98.3%. The mass capacity in tons/day before and after channelization at this cross section is show as the following charts. Sediment capacity in volume is also plotted in Figure 5.25 and 5.26 below. Before After d'DE MtecMRAS Rewtrs FO Mr M Wy to .d' .g $ 9 0d Mr 0. ma Figure 5.19: Mass Capacity at cross section 55 3500 m 3000 V 2500 2000 U to M 1500 u 1 1000 3 6 500 0 1-Jan 20-Feb 10-Apr 30-May 19-Jul 7-Sep' 27-Oct 16-Dec Time Figure 5.20. Sediment volume capacity at cross section 55 72 200000 M 150000 E 100000 a U 50000 0 1-Jan 20-Feb 10-Apr 30-May 19-Jul 7-Sep 27-Oct 16-Dec Time Figure 5.21: Sediment Volume capacity (m3/day) after channelization The lowering of bed level after providing of guide bunds due to scouring of the bed by high shear stress at this cross section is shown as follows in Fig 5.22. 96 —Seriesi —U—Series2 ,__ 94 E C 92 es 90 0 a `c 88 86 -f- 3500 4500 5500 6500 Station `m) Figure 5.22. Bed change at CS 55 5.4.3.2 Downstream and Upstream Effects Velocity of flow increases in the downstream cross sections up to 54.7*, which is around 2.7 km distance from cross section 55.5*. In the upstream cross section during high flood the velocity of the flow decreases in the upstream cross sections from 55.5* to 55.9*. In between cross sections 55.5* and 55 there is an increase of velocity due to the upstream length of the guide bank as shown in Appendix B. Similarly the shear stress is increased up to cross section 54.7*. Shear stress in the upstream cross section increases up to cross section 55.5* and then it starts decrease due to decrease of 73 velocity. Mass capacity increases in the downstream cross section up to cross section 54.4* which is 5.508 km distance from cross section 55. The effects of channelization on velocity, shear stress, and mass capacity for the simulation period is shown in Appendix E. 5.4.4 Channelization at Cross Section 50 Cross section 50 is located at 41.32 Km distance from cross section 55. The latitude and longitude of this cross section is estimated as 26°56'19.98'°N and 94°20'49.13"E. Guide bunds on both sides of the river cross section are proposed to construct at this nodal point to confine the flood plain from 6.8 Ian to 1.8 Km waterway. The dimensions of the guide bund at this cross section are: Length of the waterway =1.81an U/s Length of the guide banks = 1.5 *.1.8 = 2.7 km D/s length of the guide bank = 0.25 * 1.8 = 0.45 km Convergence towards the structure with I in 40 Total length of the guide bank = 3.15 km on both sides of the bank. WfthOutGB Plan: WithOutGB 3/28/2012 04 035 Legend E € M=J2OD400]0 o Grand 7s • m — BatitSta w 70 65 0 t000 7D00 3000 4000 9000 600 7000 Sfalan (m) Figure 5.23: Water surface elevation at Cross section 50 74 WthGB@50 Rev Plan: WthGB@50 Rev 3/29/2012 -.04 Nil Legend 85 EG 22Ju[200 40 000 AS 2ZJd2OG4 0000 70 01L S(m) Figure 5.24: Water surface profile at CS 50 after channelization There is an increase of 11.4% in velocity, 19.5% in shear stress, and 38.3% in mass bed capacity during high flow periods at this cross section after channelization. Water surface elevations, velocity, shear stress, sediment mass capacity through the simulation period before and after channelization are shown in the following plots. Before Channelization After Channelization d. M"WMS ongthedn9 Ie.d,'RS Re.4t.nd t -7 OILRA I •------. .. -. .-.. .-,------. . F Y. A. AAl, ? TW. Figure 5.25: Velocity time series before and after channelization @ CS 50 75 Before Cahnnelization After Channelization d k ------t _._. 1_I -_-- I. ------Wn I & ------Figure 5.26: Shear stress time series before and after channelization @ CS 50 Before Channelization. After Channelization di Ut.thRAS R.IM.d SlI.do.WIt.OS5OR.dO3 - -.-.-.- - -. ------:- - -- - - ---i--- -- - Tn Figure 5.27: Sediment Mass capacity before and after channelization at CS 50 54,5 Channelization at Cross Section 45 This cross section is located approximately at 26°4836.90"N and 93'51'46.85"E. At this cross section 20 km flood plain is reduced to 2.8 Km waterway to guide the flow to a specific main channel and to avoid the braided pattern of the flow. 76 Figure 5.28: Cross Section 45 Preliminary dimensions of the guide bunds at this cross section based on the guide lines are: Length of the Waterway = 2.8 km U/s Length of the guide banks = 1.5 * 2.8 = 4.2 km D/s length of the guide bank = 0.25*2.8 = 0.7 km Convergence towards the structure with 1 in 40 Total length of the guide bank = 4.9 km on both sides of the bank. 5.4.5.1 Effect of Channelization at Cross Section 45 Water Surface Elevation: The change in flow parameters during high flow periods like water surface elevation and velocity of flow are analyzed before and after channelization. It is obvious that there will be an increase of water surface elevation in the upstream part of the cross section when there is a river training work. Due to the construction of 2.8 km waterway guide bunds at cross section 45, the maximum change in water surface elevation is 1.26 m at cross section 45,4* at a distance of 5.712 km from cross section 45 during high flood. 77 —s—WS EIevAGB -4—WS EIev BGB 88 86 LIJ 84 78 44.5 45.5 46.5 47.5 48.5 49.5 Cross section Figure 5.29: Water surface Elevation Change Velocity Change: 63.63% increase of velocity is observed at cross section 44.9411 * due to channelization at cross section 45. This maximum change of velocity occurs at -918 m distance downstream of cross section 45. The change in velocity upstream and downstream part of the guide bund is shown in the following graph. ~veiocreyaua -M°VAGB 2.5 2 N S.5 U .21 C, 0.5 0 -I- 44.5 45 45.5 46 46.5 47 47.5 Cross section Figure 530: Velocity change upstream and downstream section from CS 45 Through the simulation period there is an increase of velocity at this cross section. The change in velocity is around 42.6% during low flow season and 60.5% during high flow season. The velocity during the maximum flow at July 22, 2004 increases from 0.72 m/s to 1.82 rn/s at cross section 45. The graph on velocity through time is shown in Fig 5.31 below. 78 Shear stress The maximum increase of shear stress observed from the results is 77.14% due to channelization at this cross section during high flow period. The change in shear stress and velocity through the simulation time is shown in the following plots. Before Channelization After Channelization 4DEMt&R4S iWE MC4iR.ES RI%o.,d ASr 5ROCCodO3 I.— - .1 Loguid Vdodty(im) II a Co Fab MW AM May .Aa, AA ALg Sp 08 Nw D Jan F& Mw Ap, May Am Ma.g Say 04 Nw D WM Tine Figure 5.31: Velocity and Shear stress plots at CS 45 Sediment Transport Capacity: At this cross section the sediment mass out is higher than the sediment input due to channelization. There is erosion of bed due to increase of shear stress. The sediment mass capacity at cross section 45 is shown in the following Figure 5.32. Before Channelization After Channelization iDE MtacI.S RinaC. .,d .insErna.oainet,oaca.. 4DEMMahAS- - R-4.- andi4tanUMe4RthOB©45R0.a.ady3- - - ...... a Figure 5.32: Mass capacity in tons/day at cross section 45 79 5.4.5.2 Downstream and Upstream Effects Increase of velocity and shear stress extends to 2.88 km downstream of cross section 45. After this distance the values of shear stress and velocity will be the same to before channelization of the river at this cross section. Similarly the mass capacity increased due to change of channel width in the downstream cross section up to cross section 44.8235* which is at a distance of 2.88 km from cross section 45. In the upstream cross sections, the increase in velocity and shear stress extends up to cross section 45.3333*. When we go to more upstream from cross section 45.3333* there is decrease of flow parameters like velocity, shear stress and mass capacity due to channelization at cross section 45 and further upstream cross sections have the same values as before channelization. Graphically the velocity and shear stress time series at selected cross sections are shown in the Appendix C. 5.4.6 Channelization near Tezpur At this site there is an existing bridge which is called Kolia Bhomora Bridge. The Kolia Bhomora Bridge is a 3 km bridge over the Brahmaputra River that connects Tezpur on the north bank with Nagaon on the south bank. The bridge is located around at 26°36911" N and 92°51'23" E geographically. It is one of the longest bridges in India and stands unrivalled in its grandeur. A flow guide bund on the left side of the cross section confines the flow and flood plain of the river. The bridge is located at cross section 35.3125* which is around 4.7 km upstream of cross section 35. Figure 5.33: Seattleite images at Cross section 35.3125 The dimension of the guide banks depending on Galas Guide lines from table 1 above is calculated as follows. Length of the waterway =3 km U/s Length of the guide banks = 1.5 *3 = 4.5 km D/s length of the guide bank = 0.25*3 = 0.75 km Convergence towards the structure with 1 in 40 Total length of the guide bank = 5.25 km on left side of the bank. 5.4.6.1 Effect of Channel nation at Cross Section 35.3125* Water Surface Elevation There is an increase of water surface elevation due to the construction of guide bund to confine the flow in the upstream cross sections. The maximum increase of water surface elevation is 0.32 km at 22 Jul 2004 at cross section 35.625* which is 3.8275 km distance upstream of cross section 35.3125*. tBGB -*—AGB 68 E 67 a, Ui 66 65 64 63 4- 35 35.5 36 36.5 37 37.5 38 Cross section Figure 5.34: Water surface elevation Change on 22 July, 2004 81 Velocity Change The increase in velocity in the upstream and downstream cross sections due to the construction of guide bund on the left side at cross section 35.3125* is shown in the following Figure 5.35. —4—Seriesl —U—Series2 1.6 1.5 1.4 .. 1.3 1 E 1.2 U 01 0 > 0.9 0.8 0.7 0.6 34.5 35 35.5 36 36.5 37 37.5 38 Cross section Figure 535: Change in Velocity at different cross section Velocity through the simulation period has increased up to 38% due to the construction of guide banks at cross section 35.3125* during high flood time. Before Channelization A a- EMtIRAS RmIA air Si,tai y"— LdI - _Vdty(W) - - I ---• ------ ------ - - -.- - Jae FthF& Apr May Jm Ad Aç Sep i No D Tm Figure 536: Velocity series at cross section35i125* 82 Shear Stress Shear stress at this cross section has been increased by 51.5% due to construction of guide bund on the left bank of the river. After Channelization dDEMtacIdAS I Fab Mar Apr May Jun hd AtU Sep Oct Nun Dec flm Figure 5.37: shear stress series at cross section35.3125* Sediment Transport Capacity The result of sediment mass capacity shows that there is an increase of 80% in mass capacity during high flow time of the simulation time at cross section 35.3125* due to the guide bund on the left side of the river. Sediment mass capacity at this cross section in tons/day is shown as follows. Before Channelization After Channelization UaRAS R,ad.R.c8r.J.unnwtouna...un dADEUb..HRMR—ft.W - T! ------it UW At A AC W 04 , Tm Figure 5.38: Mass Capacity in tons/day before and after channeiliation 83 5.4.6.2 Downstream and Upstream Effects Shear stress, velocity, and sediment capacity changes in the downstream and upstream cross sections are compared before and after constructing of guide bund. It is observed that there is a change in velocity and sediment mass capacity up to cross section 35.125* which is around 2.87 km downstream of cross section 35.3125*. The plots of velocity, shear stress and sediment mass capacity through the simulation period are clearly shown in Appendix D. 5.4.7 Channelization at Cross Section 29 At this cross section the waterway is tried to decrease from 20 km to 2.6 km. Estimated geographical location of this cross section is at 26°2655.68"N latitude and 92°1334.44'°E longitude. Figure 539: Satellite images at Cross section 29 5.4.7.1 Effect of Channelization at CS 29 Water Surface Elevation Change Due to channelization at this cross section the water surface elevation rises in the upstream part of the cross section up to a certain distance. Graphically the afflux water elevation during high flow records in the simulation period is shown in the following plot on figure 5.41. The maximum change in water elevation occurred in summer season is 0.48 m at cross section 29.5454* which is located 5.847 km upstream of cross section 29. 84 60 I —4--WS Elev BGB -il—WS EIevAGB i E 59 a+ 58 55 54 28 29 30 31 32 33 34 Cross section Figure 5.40: Change in Water surface Elevation at different cross section Velocity and Shear Stress Change The change in velocity at different cross section in the upstream and downstream part of cross section 29 is shown below graphically on Fig 5.41. This velocity change graph is during high flow time at 22 July, 2004. 1.6 —4--V BGB --V AGB 1.4 1.2 N 1 ' 0.8 0.4 0.2 0 28 29 30 31 32 33 34 35 Cross section Figure 5.41: Velocity change at different cross section in July 22, 2204 profile The increase in velocity and shear stress at this cross section is estimated as 72.7% and 88.6% respectively. This increase in velocity and shear stress causes high rate of sediment transport capacity at which erosion of bed material at this cross section will be maximum. 85 Before Channelization After Channelization dDEMtreh.4S Rreftaa,d d DEMtreMR.S Rre1t ata &reJ RatS WhGSRreredS3 &—talm am, 14 IS Q28 C - I 1 1 1 I F' Mar V May .An Ad Au SC) Oct Nw Dre Fth Id Aug Sep Oct Na Oat 21fl1 Mar ykm Tire Tee Figure 5.42: Velocity before and after ielization at cross section 29 Before Channelization After Channelization dTEMtatHP.AS RreAs aWsimLiatiorn%RwSirnLdaiorAWbOLLGEL~dLI3 4OEMtatMRAS RredM atd relaIm (14 Legnd lrrrrTTfl r T,T iAl i I kil I ISl1atrSrere(pa) Fth Mar Ap, May Jan .14 Aug Sep Oct Nov Oat lire Figure 5.43: Shear stress at cross section 29 before and after channelization Before Calihelization After Channelizaion dDEMI,.IC)AS Real. .,d Figure 5.44: Mass capacity at CS 29 before and after channelization 86 5.4.7.2 Downstream and Upstream Effects It is obvious that there is an increase of shear stress, velocity and mass capacity up to a certain distance in the downstream cross sections due to channelization of the river and decrease of these parameters up to a certain distance in the upstream cross sections. Due to channelization at cross section 29 the increase of shear stress, velocity, and mass capacity in the downstream cross sections extends up to 2.78 km distance from the point of channelization work is placed. The plots of each parameter through the simulation time at different cross sections in the upstream and downstream cross sections are listed in Appendix E. 504,8 Cross Section 22 (Guwahati) At this cross section there is an existing bridge called Saraighat Bridge, which connects north and south banks at Saraighat. This cross section is located geographically at 26°10'23.851° N and 91°40'20.30" E. The bridge has a width of around 1.5 km length. Saraighat Bridge is constructed over the Brahmaputra, also called the Red River, is the first rail-cum-road bridge. It was opened to traffic in 1962 by the Prime Minister Jawaharlal Nehru. This is a double deck bridge with a national highway on top and railway tracks below, Figure 5.45: Satellite images at Cross Section 22 The simulated values of velocity, sediment mass capacity and shear stress at this cross section are shown in the following plots. 87 dDEM1\RS Rofls aid M,.Atio—%RwSinUd~.IAVUBIGB~Mzm — I r--1- Logid Feb Mar Apr May Jun Jul ALQ Sop OctNov D Figure 5.46: Velocity at cross section 22 Figure 5.47: Shear stress at cross section 22 dDE MtovNR.vS Rat Figure 5.48: Mass Capacity at CS 22 88 5.4.9 Channelization at Downstream Cross Section of Guwahati As shown in the Fig 5.51 below there are two channels on the left and right bank, of the river which makes the river to have braided pattern downstream of Guwahati (Cross section 22). The satellite image of the river at two different times indicates that the channel on the right side has advanced by eroding the bank of the river more than 750m. By closing these secondary channels the flow of the river will come to flow with a single main channel. These secondary channels can be closed by river training structures like bed bars porcupine screens, jack jetty, permeable spur etc. Figure 5.49: Satellite images downstream of Cross Section 22 Velocity, Shear Stress and Mass Capacity Change At Cross Section 21.6*, 21.4* and 21.2* There is a change in water surface elevation, velocity and shear stress after channelization works have been provided at cross section 21.6*. The analysis is done by the interpolated cross sections between cross section 22 and 21. From plots of velocity time series it is observed that there is an increase of velocity due to river training structures (closing of secondary channels). 15%, 22.15%, and 17.8% increase in velocity at cross sections 21.6*, 21.4*, and 21.2* respectively during high flow periods are observed due to channelization. Similarly it is observed that there is an increase of shear stress by 16.6% at cross section 21.6*, by 30% increase at cross section 21.4*, by 25.5% at cross section 21.2* during high flood season particularly at 22 July, 2004. 89 Mass capacity increased by 87.5% at cross section 21.6* and it increases in a similar way to the other cross section up to a certain distance in the downstream part. The plots of each parameter at different cross sections before and after channelization of the river are listed in Appendix F. Further downstream of Guwahati becomes highly braided due to formation of several subsidiary channels. Channelization of these cross section using flow guide bunds will be impossible due to economical constraints and complexity of the river. The other option to channelize the river proposed is by using obstructions like Geotextile tubes, hump or bed sill and other training structures. These will change the pattern of morphology of the river through time by triggering sedimentation in the downstream part of the river. The morphology of the river is changed through time at cross section 19, 18 and 17. The two satellite image indicates that the river comes to the left channels as shown in the following images. This indicates that closing or affecting the channels on the right side of the river will change the morphology of the river. Figure 5.50: Satellite images downstream of Guhawati From the above images river training structures have to be constructed on the right side of the bank upstream of cross section 20 to close secondary channel. These channels can be partially closed by phase at different seasons by using Geotextile tubes. These geotextiles tubes protect the flow of water on the right side of the channels by allowing sedimentation on downstream cross sections of the channel. 504010 Cross Section 9 (Jogighopa) The other natural nodal point in which the width of the river is narrow compared to others is at cross section 9(Jogighopa). There is an existing rail-cum-road bridge called Naranarayana Setu constructed to connect Jogighopa with Panchratna. The length of the bridge is 2.284 km. A guide bund is constructed on both sides of the river bank at this cross section. Figure 5.51: Satellite images at Jogighopa The velocity, mass capacity and shear stress plots through the simulation period are shown in the following plots. iDEkKaS 1/r Apr U., An Jd A, 8, 0 Figure 5.52: Velocity, Shear stress and Mass Capacity Plots at cross section 9(Jogighopa 91 6o NUMERICAL SIMULATION ON THE EFFECT OF SUBMERGED BED SILLS ON RIVER 601 Introduction An obstruction of bed sill/hump on the river bed has an impact on the sedimentation system on the upstream and downstream part of the river cross sections. A 2 m, 3 m and 5 m bed sill obstructions with 400 m length towards the flow from the bank of river has been taken to analyze the behavior of flow parameters and bed characteristics. For 2 m height of hump at cross section 22 the results of simulated values of different parameters like velocity, shear stress and bed change are discussed in following sections. 6.2 Effect at the point of Submerged Bed SDI Water surface elevation At cross section 22 where the obstruction/submerged bed sill is placed, there is a decrease of water surface elevation. A change of 1.24 m is obtained during high flows due to the above bed sill obstruction during monsoon season. Velocity There is an increase of velocity at the point of bed sill and upstream cross section due to turbulence of flow. It is observed that there is an increase of mean velocity at this cross section up to 43.88 % during monsoon season. The plot of velocity time series before and after the bed sill obstruction is shown in the following Fig. 6.1 below. ¢wE umnwne R..ab .Imdana.ruaeaU(YSM O mwum®azlanymoa I- SinM kn r 1 1 1 ITI f1~~ Mar Ap May ,gym L !4q Sep O1 N. Dm Figure 6.1: Velocity at cross section 22 before and after placing bed sill 92 Stream Bed change Due to bed sill obstruction on the right bank of the river at cross section 22, there is more sediment transport capacity and lowering of bed levels on the opposite bank. Lowering of bed level up to 4.2 m after obstruction is observed at station 2075m on cross section 22 in one monsoon season as shown in the following Fig. 6.2 and Fig. 6.3 which are sediment cross section bed change at different times before and after obstruction. Figure 6.2; Cross section bed change at different times before placing bed sill Figure 63: Cross section Bed change at different times after placing bed sill 6.3 Upstream and Downstream Cross Section Effects There is increase of velocity about 30% at cross section 22.0526* during monsoon season. This cross section is in the upstream cross section 480 m apart from cross section 22. Increase of velocity causes to transport more sediment to the downstream cross sections 93 rJ o. QJ: in Feb PM, Apr PMy A. .kAugJIH' S.p O -w D 7k. Figure 6A Velocity at cross section 22.0526* before and after placing bed sill At cross section 22.1052* which is 960 m apart from cross section 22, there is a slight variation of velocity as shown in the following plots. &ME M.,tRS R..M U..M1.S R.U. 1.6 1.2 I ::Jn Fth Mw Apr My J. .UMq &p OctNo, D Th Figure. 6.5: Velocity at cross section 22.1052* before and after placing bed sill Bed change at cross section 22.0526* indicates that there is more degradation of bed after providing hump on the bed of the river. This is due to increase of turbulence flow in the upstream sections. There is a degradation of 0.64 m more after placing bed sill obstruction during one year simulation period. Figure 6.6: Bed change before Obstruction at cross section 20.0526 Figure 6.7: Bed change after Obstruction at Cross section 20.0526 At cross section 21.9411 *, 480 m downstream of cross section 22, there is an increase of velocity up to 26 %. t' U r'Rns n..a• TmivwwaIrwFOS.~rvcuw~0~l~t..aoa vradMmh~ iI I € i Z J. F M. M. Jug 31 ASV Sq OC N. D¢ k- TY~o Figure 6.8: Velocity at cross section 21.9411* before and after Obstruction Increase of velocity due to obstruction cause degradation of bed. But it is less than before placing bed sill obstruction; since more sediment are transported from upstream cross sections and the 95 sediment mass change is higher at this cross section when it is compared before bed sill obstruction at cross section 22. Figure 6.9: Bed Change at cross section 21.9411* before placing bed sill Figure 6.10. Bed Change after placing bed sill at cross section 21.9411* The next downstream cross section 960 m from cross section 22 shows more aggradations of sediments as shown in Figure 6.12 and 6.13. This is because the eroded sediments starting from upstream cross section and some parts of downstream cross sections from 22 gets deposited due to lower velocity. The following plots show the cross section bed change at different time during the simulation period before and after placing bed sill/hump. 96 Figure 6.11: Bed Change before bed sill at Cross section 21.8823* Figure 6.12: bed Change After bed sill at cross section 2L8823* Increasing the height of bed sill hump triggers the change in velocity, mass capacity, bed change and other parameters in the upstream and downstream sections. But the effect of each parameter will decrease when there is an increase of hump height beyond the optimum value as shown in the Fig. 6.13 below. Different heights of bed sill hump have different effects on the bed change of the channel, The aggradations and degradation effects depend on the height and width of the hump used. For 400m width and for different height of hump the following aggradation at 960 m downstream of cross section 22 and degradation at cross section 22 are plotted as shown in the following Fig.6.13. 97 -4-22 _4-21.8823* — 6 $ _, 5 0.8 E N N 4 0.6 00 aJ3 10 c02 0.4 1 0.2 C ®0 T 0 0 1 2 3 4 5 6 Height of Hump (m) Figure 6.13: Optimum hump height 98 7o SUMMARY AND CONCLUSIONS In this study numerical model has been developed for Brahmaputra river for the whole reach in Assam which covers 622 km. A quasi unsteady flow and sediment transport analysis has been done by using HEC RAS. Hydrological, morphological data, sediment data and temperature data for four years of simulation time from 2004 to 2007 have been used for the development of model. The second study includes channelization works at selected nodal points and their effects on flow parameters, like velocity, shear stress, sediment load and change in invert elevation have been analyzed. From the study the following conclusions have emerged. From calibration of roughness of the river cross section it has been observed that there is seasonal variation of Manning's coefficient. There is a decrease of roughness coefficient during monsoon season and increase of roughness coefficient during lean season when water surface elevation is decreased. During Calibration of sediment transport function it is observed that Yang's method of transportation gives good result than Ackers-White and Engelend-Hansen method of transportation. Especially Ackers-White method gives high result of sediment load. e It is observed that there is change of bed elevation during the time span of simulation period. Aggradations and degradation are observed at different cross sections. For example aggradations are clearly observed at cross sections 53, 50, 46, 38, 36, 32 4 and others. Similarly at cross section 54, 51, 47, 41, 39, 37, 33, 22, 21, 9 and others there is degradation of bed. s Channelization works along Brahmaputra river are simulated at some nodal cross sections. ® From channelization it is observed that there is an increase of water surface elevation in the upstream cross sections up to a certain distance based on the type of river training structure used. There is an increase of velocity of flow, shear stress at the bed, mass capacity of sediments at the point where channelization work is provided and downstream cross sections.- The maximum increases of velocity at maximum flow are 32%, 66%, 63.63%, 38%, and 72% at cross section 57, 55, 45, 35.3125* and 29 respectively. Similarly the increase of shear stress at this selected cross sections ranges between 49 to 88%. 0 Providing hump/raise of river bed will change the morphology of the river by sedimenting the downstream cross sections. o Closure of channels on numerical experiment has been indicated that there is degradation of cross sections in the upstream cross sections and aggradation on the downstream cross section. 100 ANNEXES Appendix A: Downstream and Upstream Effects of cross section 57 1. Downstream Velocity before and after channelization from cross section 57 Before channelization After Channelization d1DEMtsMR/l5 Rmdm addmlabrtdta.5.nlctloAW hQRGEUadJ7 SDE YE¢NR49 RwuftoAWLdtim%iwSIMUBWMYAIhW@67Rna dW ~...... _.. t Vdadty(n) F I 3 Vtl dlyy 14 -.._. . . _ _ ___ a I- {.. . ryr to 3 a lD Ln z i .ttn AM 8aP &t N. D Jar Fda Mar Ap May .Mm .w AL, Say Od M- Ds TYrn Ti,. LDE1aO1$P SRmtamddmlatlwatRarSYnddfvl,W mm3 6'DE 6{twtAR/3 RaLLb ~q d11t1d5onV iw BY: W dbbYAt:C67Rnae665 5mlaEm j Si ue9 i ~ Vdadty(m~s) 11T 00 00 i I 00 i € j 00 I~ as . '. _._.:_ __ __...... _ - .<:__..• _ In Ma AW May An .0 fl/40 Say Oct-1 NW Jut F b Mar AR May Jun .Md Mp Say Od Non Gm arr Ti,. Tha 4 EUt.MRAB Raub ,o elmldbnWwBkro O *Y4NG67Rw.o01 Srnbim Vdity(rtIe) i F i Lir ) Rb Mw Ap May ,kn A9 Say Od MW Dc Ft Mm Ap May .Yn .1d /yy T7.. Stan 101 2 Downstream and upstream shear stress before and after channelization at cross section 57 Before channelization After Channelization 00 in to U) 00 CO 00 CO d Ln 102 dWEtrG,WRem!a .JWW elRmr9~ WOOZWB m m m Co tD un „ „V m 3. Downstream and upstream Mass Capacity before and after channel nation from cross section 57 Before channelization After Channelization Im m 00I 103 00 00 00 00 l0 Ln m m m Co 104 Appendix B: Downstream Effects of Cross Section 55 1, Velocity Before channelization After Channelization RIDE YEa11AA4R®b etl tlmld5~Y2w9Y:fl~uiWPO:I(:8w® C`DEYt.CIER 9R1.nEYmdYla:~Ww3h:U~bMMRnae®Mnzwem Snutilm ~~€ Si,',',. l_ E € i ; Vda j VaotlMmbl r i € y to i....._~-__•._._.~_ ...... f ...... ~ ...i ....._. ~f a ~ : c Ln a.hn F® Ys M MW L: .1A h0 S, Od Ib Di a J.. F® Y> AR IOoy An d/ AVI3 EV Od Na OW T' TFo Ti,. QY1E Y1sMPh4 Rm4 ad tlmiglonVtw SiNas>tiW IrOLLW ~a~ OWE U0IERA9 Rsub tldmdflbi,R.. ,ddbth O6®'dfa..d61 67m1almE H sRy1m~:-.~.._...~._:..,— to —~ ~ • ~ —1 Lgme , I E j 1 11 Ll H'Tt Mar M M" end &a 8,W Yav Dc den F~ Yr A,' 1by o^ 1 AW 6fiS ad IOM t- Ti TY1e 6~DE YOC W i~$ Rw,p eq tlmlab W 1w5YNdf n~W IAU(ianea57 dWEE 0fhRA9 Rm:Abdmltlb:wSi3O EWM0006R,.10 _ _ _ __~€ Lgcfd i f \' 1 i 3 a6 SE Od Rar Dc .6n FO Yar M Yof.1.., f art Na 00 T" 105 2 Shear stress Before channelization After Channelization Ln Iq Ln U, 106 dVE MOOR* J77 LCed hL Ln AM 1 1 .. MU Th. 3. Sediment Mass capacity Before channelization After Channelization tt LI'h Ln - I 4 * Cn 14 Ln MM 107 - * 00 Ln Tft a N Ln K 108 Appendix Co Downstream Effects at CS 45 1 a Velocity Before channelization After Channelization ,ME MWHRAS gwoe,a IkM CW*"Sb—v wan:v OE~ t ! - vtl~71a+~1 Vemry(Mel 1 ' t .._ ...... _..._i_.__..... `.....~...... '+:..... __. 0) i ' 3 b: F~ Ib b kw1 .1n ~+9 SG W Ib no °b: Fa u:bi mA:a BV oa ramD Ti 4WE ktrM R.:.bmNtlnYitrNFts,Switla{YN~I®n® aO uy.upws rtm Mum.au.wwe~narnaANn08p~nr.~.~af 6YM1l fm ~~ i i hhII i vtlaaM:tll c) :1111 3 N co co I a j--.i { a M. . A:¢ $q o Nw/ Oi Jm Fa M. b Lb/ do J9 :Ya SW m N ac .b: Fa Ird ~P V9' K WM WASRMedehnlibawrsNCMNSm~¢¢a~ a:OE utcmns X.. b watln:lanw8 nSSAMlGOC iruaW a .__..._.._~...... _.' 5✓.:e1~1ai vtl7fan} vewt~y:ny) i i e i 1 } I j I• Ap Yq .1n •Ma BT Oe Nw as .r: Fm i 44 An A:v SP of Mor D® Tlm 11 109 2. Shear stress Before channelization After Channelization QIDE YYennws Fmb.aon.~eanw~S:n uronw,F:oB®au...dm i ' 1 ' i e 1 i Slobse4lW g i in Fb Ym Aw 1!F . , .0 Mp Sp Oy Na Do SIDEYtcrgb Nb.NOm:t.a.,sb.4 bwto6®4yq.,. .J ...~...... SiM1b9m ... 1 e ; Jn FSdI ,AR Lft An J A 6p as Ib Da Tk. G'OL YtcN113Ndb em1~m1Rb9YnY~iWMiI:OB®45a~uEW Sidd- 1 E g 1 jti 111 1 - ~f @ I ~ 1 . Fa . M W .ln .0 A:G Bq as Na Dn i' 110 3. Sediment Mass Capacity Before channelization After Channelization LLD ENItP RA9 Rm6 .W mfl. S.rniS MMAT09@URlits~lQi 0) v SO Od Nd Dc J:DEMt PtPA9 Rdb edSs,jd ~WrSYnJ/uR~MUC8® 111 Appendix Do Downstream and Upstream Effect at CS 3503125* 1. Velocity Before channelization After Channelization w EL ..Fb Ir M J. Jn JI IYY sv Oe Ner Om avr 7ir. At0LG11R16 Rsbod~mldb.~lierSinllioiWM6Sv~ 9ndim ~~9 N a M moeW R G R .da s:~.uw,a Ga.® Y Sim ~i+l . Fa M. M }h' Jn J" A N. Go TYw dM&R.(W8 R~bmd~mlItrnlRrSF.fdoddIO U, a S en 112 2 Shear stress Before channelization After Channelization 11 10 I LA U .. 113 3. Sediment Mass Capacity Before channelization After Channelization a1LEYu xRASR wun:om~abnWrtm.e IdM*WIGBA]LA] JT Z ~ ` N m Z 3 ! ? m Ln m r Ee Ro b W M r M N u~ a o: 6~OE YI~tNAl4 FYUN.~eumlaw.~RrB~mILb~KtnOBO]6s~] QNEYspJUs RSef d4lv~WrSI,d* t hOBO11. 3 * U, N 00 U m i r ro W N W r u M Yl OC iw oR iNEYer,YUS Rruh.W d~.IavI31w BhWrn~Cx~OBC>~.4f O~Y~w a U, TiN mLn 114 Appendix E: Downstream Effect at Cross Section 29 L Velocity Before channelization After Channelization A. 0 rn 0 rn CO 00 N 115 2. Shear stress Before channelization After Channelization Ft- A. W7J .. AM $m Otl T- cq co N k. b My A.. .0 Ao 9a oa P N Ip 1* J' .•• kc Fp o. 116 3. Sediment Mass Capacity Before channelization After Channelization a10E UtcM11A5RdbsMC:nl6b.ar8enJ~el'vM1TOJGfl a~ ~WrA f ~ S ~ e i i s ab E utc ~W A4 Fo:u .a e-:nleaYA r 5naaienvf nOJCd~a® . 1 ~csh Fe.vy~ } co 03M NW ti ¢~DEl WMRASR. nCumlgbtRe&rnit R•ITOIG9~® C a : i C..1 y i N ti CD N 3 1 f 117 Appendix F: Downstream Effects at Cross Section 21.6*, 21.4*, 21.2* and 21 1. Velocity Before channelization After Channelization d EELOSHRAS Rstastltlml®ueWr5m1[b,aNfQtOflw® d,OEYcMRd3R= atltl W,U& ?NRr6"afixi AfGe@ "d33 Sinitle H - L mN L c:d a' I in F. !Vr Yyr }n Ad Mp 5® OR Mmr. Dm Jn Fm LRf up Yq Jn .fl A:y 8y Od Na Dm Th. T d=AM.tUW R~oe:ds. dCb P-3S M,W,PeGB.wEO) QWEL~?Wi5 Rsbrtl~Wr6M~Fi:fGB®11.ed7J ....._i.... Vaady(Rh) ! ! I i I i_• I i j 1 ic N as.....i....._€...- ---:...... _ — - - J. Fd MW M LW Jn M &9 By, Odt Mmr Dm A! Fa Rb b Mb Jn LAO M Od N. Ds lb. TBv d10E us b.i ttMIAeGa..m d:DE fsHRll9 R®b N dm1~Y~Lr5N W ~eHYAf~C ~ ..~l 1 —^ _; ~ i : vtlmtYlmq I i t _ I y ' at - . , Fd I .0 Jn d b BOW M Oc .kn Fm W M Lb7 .b Jd k S, Cl Nn Dm lb TFa 118 2. Shear stress Before channelization After Channelization cD cm U. aWEArWLW amje.~a~~.~.w..~.e~nx~.~uas.® 129 3. Mass Capacity Before channelization After Channelization csJ 120 Appendix G: Downstream Reach Length and Distance of Cross Sections of Brahmaputra River Distance from 39 40 383.03 11.22 Cross Dn base Cross 40 41 389.66 6.63 S No. Section Distance section 41 42 398.33 8.67 1 2 17.34 0 42 43 412.09 13.76 2 3 28.05 10.71 43 44 423.31 11.22 3 4 38.25 10.2 44 45 439.63 16.32 4 5 46.92 8.67 45 46 453.91 14.28 5 6 56.61 9.69 46 47 465.13 11.22 6 7 66.3 9.69 47 48 474.82 9.69 7 8 73.44 7.14 48 49 483.49 8.67 8 9 82.62 9.18 49 50 490.63 7.14 9 10 92.82 10.2 50 51 498.8 8.17 10 11 100.98 8.16 51 52 505.94 7.14 11 12 109.65 8.67 52 53 513.08 7.14 12 13 119.85 10.2 53 54 522.77 9.69 13 14 128.01 8.16 54 55 531.95 9.18 14 15 137.7 9.69 55 56 541.13 9.18 15 16 146.37 8.67 56 57 558.98 17.85 16 17 156.06 9.69 57 58 570.2 11.22 17 18 167.28 11.22 58 59 579.38 9.18 18 19 175.95 8.67 59 60 589.07 9.69 19 20 182.5 6.55 60 61 601.82 12.75 20 21 189.21 6.71 61 62 613.04 11.22 21 22 197.37 8.16 62 63 626.3 13.26 22 23 206.55 9.18 63 64 634.46 8.16 23 24 213.18 6.63 64 65 640.07 5.61 24 25 218.79 5.61 25 26 224.91 6.12 26 27 234.6 9.69 27 28 241.23 6.63 28 29 251.95 10.72 29 30 262.15 10.2 30 31 272.35 10.2 31 32 284.08 11.73 32 33 296.83 12.75 33 34 310.1 13.27 34 35 325.9 15.8 35 36 341.21 15.31 36 37 352.94 11.73 37 38 365.18 12.24 38 39 371.81 6.63 121 Appendix H: Simulated and Observed Water Surface Elevation Using Different Method of Transportation 1 Cross section 22 (Pandu) A0kWNt 5O 4. - ISDOD000 1/1/2004000 79/2004000 2/20090.90 0/22/2000000 4/20/2006000 11/26/2006000 6/14/2007090 r.,,,A4.l 51 41 kj 41 41 II II! 30 1/20040.90 7/29/2004090 2/24/20050.00 9/22/20050.00 4/23,2009090 11/16/2000 2.03 6114/20670.00 1/2/2909000 1/23/2094000 2/24/2005000 9/22/20690.00 4/20/20290.00 11/16/2006000 6/14/2007090 vk-(D.M 122 2. Cross section 9 (Jogighopa) 0001,9-WPütO 40 -ShrAWJ 34 28 -S--W 4A 25 1/1/2004000 - 7129/20040.00 2/24/2005000 9)22120095.00 4/10/2004000 11/16/10040.00 0)1412007000 40 OLA 34 pim MIR+- za 1/1/2004000 7/29(2004000 2/24/2005000 9/22/2005000 4/20/2006000 11)14/2006000 6/14/2001000 05- .49 -054DI) 34 28 1/1/2004000 '/29/2004000 2/24/1005000 9/13/2005040 4/00/20060.04 11f16/2009000 5/14/2005000 123 BIBLIOGRAPHY A.Gyr and K. Boyer. (2006). Sediment Transport A. R. Zarrati.etal (2005). Mathematical Modeling of Meandering Channels with a Generalized Depth Averaged Model Arora, Dr. (2002). Irrigation, Water Power and Water Resources Engineering. B.prezedwojski. (1995). River Training Techniques. Baosheng Wu et al, M. (2005). Case Study: River Training and Its Effects on Fluvial. Ben Chie Yen, F.ASCE (2002) Open Channel Flow Resistance Chang, H. (2008). River Morphology and River Channel Changes. Chang-Lae Jang and Yasuyuki Shimizu. (2005). Numerical. Simulation of Relatively Wide, Shallow Channels with Erodible Banks Coleman, J. M. (1968). Brahmaputra River: Channel Processes and Sedimentation. David Shankman (1991). Channelization Effects on Obion River Flooding, Western Tennessee Douglas G. Emerson, A. V. (2005). Evaluation of Drainage-Area Ratio Method Used to Estimate Streamflow for the Red River of the North Basin, North Dakota and Minnesota. Eric C. Carson. (2005). Hydrologic modeling of flood conveyance and iin2acts of historic overbank sedimentation on West Fork Black's Fork, Uinta Mountains, northeastern Utah, USA Hongwei FANG. (2003). Case Studies of Three-Dimensional Numerical Simulation For Total Sediment Transport I.N. Sarma. (2005). Fluvial process and morphology of the Brahmaputra River in Assam, India J.P. Bravard. (1997). Principles of engineering geomorphology for managing channel erosion and bedload transport, examples from French rivers Kenneth F. Karle, P.E. (2008). Hydrology and Hydraulic Analysis of the Lowe River near the Alpine and Nordic Subdivisions, Valdez, Alaska Kesel*, R. (2003). Human modifications to the sediment regime of the Lower. Massimo Greco, Armando Carravetta & Renata Della Morte. (2004). River Flow Olsen, N. R. (2007). Numerical Modelling and Hydraulics. Pierre Y. Julien (2002). River Mechanics 124 Ri Garde & K.J. Ranga Raju . (2000), Mechanics of Sediment Transportation and Alluvial Stream Problems R.J.Garde. (2006). River Morphology. T. S. Strelkoff, M.ASCE, and A. J. Clemmens, M.ASCE (2005). Transport Capacity for Eroded Silts in Irrigation Furrows S Lawrence Dingman.(2009). Fluvial Hydraulics Steven A Kuehl .etal (1997). Subaqueous delta of the Ganges-Brahmaputra river system Soumendra Nath Kuiry and others (2010). Coupled ID--Quasi-2D Flood Inundation Model with Unstructured Grids Suiji Wang and others. (2006). Relationship between suspended sediment load, channel geometry and land area increment in the Yellow River Delta US Army Corps of Engineers, (2010). HEC RAS, River Analysis System Hydraulic Reference Manual Version 4.1,0 US Army Corps of Engineers. (2010). HEC RAS, River Analysis System Users Manual Version 410 125