- Deva K. Borah (PhD, PE) Chesapeake, Virginia, USA
Abstract: The Brahmaputra River in Assam has several constrictions as can be seen in Goggle map. Constrictions at Majuli, Shilghat, Guwahati, and Goalpara have been playing a major role in flooding and erosion of the river and have not received enough attention. A mathematical model of the river is needed to account for effects of these constrictions, as well as, designing flood and erosion control or prevention measures, especially designing and aligning an adequate and stable channel of the river. Erosion-control measures must be designed and built immediately at active bank erosion sites where properties and human lives are at stake. For sustainable solutions to flooding and erosion, the river must be carefully studied using mathematical model(s). Adequate and stable channel reaches with sufficient capacity to convey the incoming water and sediment without damaging flood, erosion and siltation must be precisely designed and optimally aligned using the mathematical model. The stable and optimally aligned channel must be built and maintained using bank stabilization technologies and minimal dredging.
Introduction
It is encouraging to see much interest in recent years by public and private organizations or groups in addressing the serious flooding and erosion problems caused by the Brahmaputra River and finding immediate relief and permanent solutions to these problems. Investigations by Coleman (1969), Goswami (1985), Sarma (2005), and Sarkar and Thorne (2006) provide valuable information and background on the Brahmaputra River. Phukan et al. (2010) compiled all the available reports on past investigations on the Brahmaputra River and documented their findings and key issues.
As outlined in Phukan et al. (2010), there are unique and complex flow patterns in the Brahmaputra contributing to its unique flooding and erosion problems, braiding and shifting of thalweg (deepest part of the river) are among those processes. In this article, two other key factors controlling flooding and erosion in the Brahmaputra, as may be observed in the Google map of the river, are discussed. The Brahmaputra River has several constrictions along its path in Assam, mainly at Majuli, Shilghat (east of Tezpur), Guwahati, and Goalpara. These constrictions have been playing a major role in flooding and erosion of the river and have not received enough attention yet. Figure 1 shows an example of the constriction at Majuli and its effects in forming braided channels downstream of the constriction. Handique (2010) discussed the constrictions at Guwahati and Goalpara causing high backwaters and flooding, and suggested bypass tunnels through the rocky hills for fast release of the impounded water.
A mathematical model of the Brahmaputra River is urgently needed for designing flood and erosion control measures for immediate construction, and also for designing and aligning an adequate and stable channel of the river leading to sustainable solutions to flooding and erosion. Mathematical models are inexpensive versatile tools to investigate complex natural processes such as flooding and erosion in a river system and find effective measures to prevent or control these destructive natural processes. Once the models are built in the computer, calibrated and validated using physical dimensions and observed or monitored data, those are run with virtual implementation of the preventive or control measures or physical alterations to the river. By comparing results from different model runs with various alternative measures, the most effective measures or physical alterations in preventing the destructions can be determined. Because the model results (e.g., flow rate, water depth, velocity, erosion, deposition, and sediment discharge) are continuous in space and time, those provide the best data for designing control measures.
In case of the Brahmaputra, designing and aligning an adequate and stable channel is the key to sustainable solutions to flooding and erosion. The designed channel must have enough cross-sectional area and longitudinal slope (balanced) to carry the incoming flow of water without causing flooding destructions. Also, the designed channel must be in a quasi-stable, if not stable, state with just enough transport capacity to carry the incoming sediment without further erosion of the river bed and bank, and deposition or siltation. The channel will be variable throughout its length in Assam in order to match or balance variable local conditions. Such variable adequate and stable channel reaches can only be designed using mathematical models.
Figure 1. The Brahmaputra River near Majuli with a constriction upstream (far right) and braided channels downstream (river flowing from right to the left)
Constrictions in the Brahmaputra River
As can be seen in Google map of the Brahmaputra River, the river has several constrictions along its path in Assam, mainly at Majuli, Shilghat (east of Tezpur), Guwahati, and Goalpara. The Guwahati constriction is a long stretch beginning approximately near Mayang where the river narrows considerably and slowly towards the narrowest constriction from Kharghuli Hills to Guwahati University. The river extensively spreads out again further west of Palashbari. These constrictions have been playing a major role in flooding and erosion in the river.
As indicated by Handique (2010), constriction creates a lake behind it. During monsoon flood, constrictions create bottleneck or flow congestion and are unable to pass the high flow at normal depths resulting in raising the water depth substantially high and creating large cross sectional areas to pass the incoming water. Such rise of water level at a constriction creates backwater effect on its upstream river reach raising water levels as far upstream as its effects cease and thus creating a lake. As the water level gets higher at the upstream lake, water spreads out to the entire river width submerging the sand bars, the islands, and its flood plains resulting in reducing flow velocities as well as sediment transport capacities. Monsoon floods carry enormous amount of sediment from its upstream drainage basin. Due to the low sediment transport capacity at the upstream lakes, a large portion of the sediment, specially, the larger particles get settled out building more sand bars or islands or enlarging the existing ones.
Different processes take place downstream of the constrictions. Due to sediment deposition upstream of the constrictions, relatively cleaner water with less sediment content passes through the constrictions with transport capacities in excess of the sediment being transported. As a result, the hungry for sediment water erodes river beds or banks whichever is found easier below or downstream of the constrictions. In case of the Brahmaputra having alluvial banks with fine silt materials, the banks become the primary victims.
A constriction may be envisioned as a narrow outlet to a fully flowing pipe where water comes out as a jet. The Brahmaputra constrictions add tremendous velocities as well as sediment transport capacities to the water, thus adding more forces to erode downstream alluvial banks. Further downstream from the constriction, as the water reaches the wider part of the river, water velocities and carrying capacities go down, and sediment particles start settling down beginning with the larger particles.
River stretches (reaches) between the constrictions go through combinations of both the upstream and downstream of constriction processes at different times. Sediment deposition primarily takes place at the middle of the river creating the islands and sand bars, thus pushing the main channel (flow) or the thalweg to the sides causing accelerated bank erosion. The Brahmaputra River near Majuli as shown in Figure 1 is a perfect example. Bank erosion continues during the dry season as the main flow (current) is in direct contact with the highly erodible banks. The pictures shown in Figure 2 are examples of such erosion during dry seasoned low flows. These pictures were taken by the author on Janury 10, 2010 at Bhurbandha, southern bank of the Brahmaputra between the Silghat and Guwahati constrictions.
 A Bhurbandha family is losing almost |
 Highly erodible alluvial bank materials |
Effects of the constrictions may be noticed on Goswami’s (1985) gross aggradation-degradation evaluation along the Brahmaputra using suspended sediment measurements at Ranaghat, Bessamara, Bhurbandha, Pandu, and Jogighopa during 1971-1979. He reported net degradation (erosion or loss of bed material) between Ranaghat and Bessamara, aggradation (deposition, siltation, or gain of bed material) between Bessamara and Bhurbandha, degradation between Bhurbandha and Pandu, and aggradation between Pandu and Jogighopa. These are gross evaluations as the measurements were taken too far apart. In reality, erosion and deposition are more variable along the river and across the river similar to flow depth and velocity. Appropriate mathematical models are capable of predicting such spatial and temporal variations of flow depth, flow velocity, sediment discharge, aggradation, and degradation.
Measures must be placed downstream of the constrictions to divert or deflect the high velocity water from the banks towards the middle of the river so that the river could naturally dig and create the main channel through the middle instead of flowing by the sides and eroding the banks. Current flow patterns at many locations, such as Majuli, Kaziranga, Bhurbandha, and Palasbari are the main channel flowing by the banks, and these areas will benefit from such diversions or deflectors placed upstream.
Developing a Mathematical Model of the Brahmaputra River
Development of a hydrodynamic (flow routing) and sediment transport model of the Brahmaputra River in Assam is urgent. The model will generate spatial and temporal variations of flow (water discharge), flow depth, flow velocity, sediment discharge, erosion, deposition, bed elevation changes, and bank erosion (migration) along the Brahmaputra River based on conservations of water and sediment masses and momentum (energy). As monsoon (wet) and non-monsoon (dry) period flow patterns of the Brahmaputra are completely different, two versions of the model, monsoon version and non-monsoon version, may be required. During monsoon, the Brahmaputra flows almost in one channel while during non-monsoon period, the river flows in various channels between islands and sand bars, mainly on the sides eroding more valuable lands and displacing inhabitants.
Mathematical models are computer programs (software) written from equations or solutions of equations (governing equations) simulating (representing) physical processes such as flow of water, erosion of land surface including river beds and banks, transport of sediment (eroded soil), and deposition of sediment (siltation). Currently, there are a number of models capable of simulating river flow and sediment transport. Review of some of the models may be found in ASCE (1998) and Garcia (2008).
Some of the available models are proprietary (commercial), such as the models sold, maintained, and managed by the Danish Hydraulic Institute (DHI). Capital, maintenance, and support costs of these models are substantial. However, these models are easy to use because of their user friendly graphical user interfaces and available support. There are some public domain models, such as the models developed and maintained by the U.S. Army Corps of Engineers. The public domain models usually come with comprehensive documentations and some user interfaces. Users with hydraulics, computer, and mathematical modeling background should be able to set up and run public domain models. Public domain models are generally free. However, proprietary versions of the public domain models are also available with friendlier user interfaces and support with associated costs. The public domain models are generally simple and may not have sophisticated simulation capabilities. Understanding the basic structures, principles, approximations, and limitations of the models are critical in successful applications of those and benefiting from their full potentials, also avoiding misuses.
Research models developed by individual researchers are also available for use, such as the author’s stream/river sediment transport model STREAM (ASCE, 1998; Borah et al., 1982; Borah and Bordoloi, 1989; Borah and Dasputre, 1994). Such models are mostly free as those are used in continuing research and development. For unique river systems such as the Brahmaputra River having unique behaviors and unique problems, research models provide better opportunity to make the models fit with the unique situations through model modifications and expansions, which is impossible with proprietary and even public domain models.
For hydrodynamic simulation or flow routing in the Brahmaputra River, a public domain model from the U.S. Army Corps of Engineers Hydrologic Engineering Center (USACOE-HEC, 2010) called the River Analysis System (HEC-RAS) may be used. Both steady-state and unsteady-state flows can be routed along the river by using this one-dimensional model. Longitudinal and cross-sectional bed-profile data of the river are needed to enter into the model to represent the river. Observed or estimated upstream and tributary flow hydrographs are needed for entering inflows into the model system, the Brahmaputra.
Using estimated friction parameters, the model must be run for various flow or flood events and predicted hydrographs compared with observed hydrographs at gauging stations along the river. The parameters are adjusted to come up with the best hydrograph comparisons, a step known as “model calibration.” Using those parameters, other flow or flood events are simulated to ensure good prediction or comparison, a step known as “model validation.” If necessary, parameters may be further adjusted and model may be further validated until the best representative parameter set is found.
For sediment transport modeling of the Brahmaputra River, the author’s modeling program STREAM (ASCE, 1998; Borah et al., 1982; Borah and Bordoloi, 1989; Borah and Dasputre, 1994) having capabilities of simulating real time river bed and bank erosion, bed armoring, sediment deposition, and graded (non-uniform) sediment transport may be used. The model source code may be modified as necessary to model a large river system as the Brahmaputra River. The model uses spatially and temporally variable water flow, depth, and velocity from the hydrodynamic model results. Similar to the flow routing model, the sediment model will also use observed or estimated sediment discharge time series from upstream Brahmaputra and the tributaries as influxes of sediments, in addition to sediment, bed material, and bank material compositions and particle size distributions along the river. Model calibration and validation procedure are the same as the flow routing model, except for using sediment concentration or discharge measurements.
Benefits of a Mathematical Model of the Brahmaputra River
Model results can be used in the appropriate sustainable design solutions of flooding and erosion. The model can also be used to study any adverse impacts downstream of the project site. It is used to evaluate alternative flooding and erosion prevention measures and designs in their effectiveness in preventing flooding and erosion, as well as any adverse impact elsewhere.
The model can be used to design and align stable channel reaches of the Brahmaputra River with sufficient carrying capacities to carry the incoming water and sediment without having damaging erosion and siltation, building of which through structural-nonstructural measures and minimum dredging will eventually bring near permanent (sustainable) solutions to flooding and erosion.
Size (flow cross-sectional area) and grade (slope) of the river channel is critical in carrying the design flow and associated sediment without causing damaging flooding and erosion. Based on conservation of mass (continuity), water flow (water discharge in volume per unit time) is the same at two different sections (locations) of the river unless there is substantial inflow from tributaries in between. If the flow velocity (speed) is higher, a smaller cross-sectional area, similar to a constriction, is enough to convey a certain flow, whereas a larger cross-sectional area is needed if the flow velocity is lower.
Grade or slope of the river bed plays an important role on water velocity. Water velocity increases with increasing slope, thus needing smaller cross-sectional areas. Vice-versa, water velocity decreases with decreasing slope, thus needing larger cross-sectional areas to convey the same flow. Sediment transport capacity increases with increasing velocity causing more erosion of river bed and/or bank.
River width and depth are important factors in flow velocity and sediment transport. For wider channel, the depth is lower to have the same cross-sectional area. That brings down the velocity as well as shear stress and sediment transport capacity, leading to siltation.
These are a few examples of the natural complexities in water flow and sediment transport. Mathematical models take into account all these and more complex dynamic processes involved in flow of water and transport of sediment in river systems. From the existing cross sections, channel patterns, and grades along the Brahmaputra River, cross-sections of a single channel can be determined (designed) by running the calibrated and validated hydrodynamic and sediment transport models in a trial and error basis with various cross sections and various alignments.
Different alignments have different grades. For example, a straight channel will have the steepest grade. A meandering channel will have smaller grade as the channel length will increase. Efforts should be made to choose an alignment along the larger existing channel away from erodible river banks so that excavation by dredging to build the designed channel will be at a minimum.
Both the hydrodynamic and sediment transport model will help to choose the grade and cross-sectional areas along the river and create an equilibrium condition where the designed channel and alignment will have the right water velocity and depth carrying the design flow within the banks (no overtopping) and having the right sediment transport capacity to carry the incoming sediment without causing further erosion or siltation.
Conclusions
The constrictions along the Brahmaputra River in Assam, such as the ones at Majuli, Shilghat, Guwahati, and Goalpara, have been playing a major role in flooding and erosion of the river. Measures must be placed downstream of the constrictions to divert or deflect the high velocity water from the banks to middle of the river so that the river could naturally dig and create the main channel through the middle instead of flowing through the sides and eroding the banks.
Mathematical models of the Brahmaputra River are urgently needed for designing flood and erosion control measures, which must be built immediately at active erosion sites where properties and human lives are at stake because protection of lives and properties are basic human rights.
The model can be used to design and align stable channel reaches of the Brahmaputra River with sufficient carrying capacities to carry the incoming water and sediment without having damaging erosion and siltation, building of which through minimal dredging and structural-nonstructural bank stabilization measures will eventually bring near permanent (sustainable) solutions to the flooding and erosion problems.
After the designed channel being built using dredging and bank stabilization technologies, the channel will have to be regularly maintained through dredging unwanted sand bars from the main channel that may form during extreme flooding events and repairing any damages to the bank stabilization structures. Without proper maintenance, all the efforts or capital costs of designing and building the adequate and stable channel will be wasted. Maintenance must be taken extremely seriously.
Implementation of the Concepts and Approaches
The author presented some of these concepts and approaches to research groups from Guwahati University, Indian Institute of Technology at Guwahati (IIT-G), and engineers from Government of Assam Water Resources Department and private consulting practices in two occasions: North East India International Meet – 2010, Pragjyoti ITA Center, Guwahati, January 11, 2010 and Workshop on “Seeking Sustainable Solutions for the Brahmaputra: Challenges and Opportunities,” December 18-19, 2010, Hotel Gateway Grandeur, Guwahati. Research groups at IIT-G under Professors Arup Sarma and Chandan Mahanta have been using mathematical models to study and design remedial measures at several bank erosion sites along the Brahmaputra River in Assam. These groups are in a process of developing flow and sediment routing models of the entire length of the Brahmaputra River in Assam where the concepts and approaches presented here would be useful. The author has been in touch with these groups providing technical support.
A newly formed group called the Core Professional Group for the Brahmaputra (CPGB) www.brahmaputragroup.org where the author is an Executive Member is actively searching for and implementing solutions to the flooding and erosion problems of the Brahmaputra River using these and other investigative and design approaches.
References
ASCE Task Committee on Hydraulics, Bank Mechanics, and Modeling of River Width Adjustment. 1998 RIVER width adjustment. II: Modeling. Journal of Hydraulic Engineering, ASCE, 124(9): 903-917.
Borah, D.K., C.V. Alonso and S.N. Prasad. 1982. Routing graded sediments in streams: formulations. Journal of the Hydraulics Division, ASCE, 108(12): 1486-1503.
Borah, D.K. and P.K. Bordoloi. 1989. Stream bank erosion and bed evolution model. Proceedings of the International Symposium on Sediment Transport Modeling, August 14-18, 1989, New Orleans, LA, ASCE: 612-617.
Borah, D.K. and M.S. Dashputre. 1994. Field evaluation of the sediment transport model STREAM with a bank erosion component. Proceedings of the 1994 National Conference on Hydraulic Engineering, August 1-5, 1994, Buffalo NY, ASCE: 979-983.
Coleman, J.M. 1969. Brahmaputra River: Channel processes and sedimentation. Sedimentary Geology, Vol. 3, 129-239.
Garcia, M.H. (ed.) 2008. Sedimentation Engineering: Theory, Measurements, Modeling, and Practice. ASCE Manual 110, ASCE Environmental and Water Resources Institute, Reston,VA.
Goswami, D. 1985. Brahmaputra River Assam, India, Physiography, Basin denudation and channel aggradation. Water Resources Research 21(7), 959- 978.
Handique, G.K. 2010. The Brahmaputra River System – Its Problems, Challenges & Potential. North East India International Meet 2010, January 11, 2010, Pragjyoti ITA Center, Machkhowa, Guwahati, India.
Phukan, A., R. Goswami, D. Borah, A. Nath, and C. Mahanta. 2010. River bank erosion and restoration in the Brahmaputra River in India. Proceedings of the ASCE-EWRI 3rd International Perspective on Current & Future State of Water Resource and Environment, January 5-7, 2010, Chennai, India. ASCE-EWRI, Reston,VA.
Sarkar, M.H., and C.R. Thorne. 2006. Morphological response of the Brahmaputra-Padma-Lower Meghna River system to the Assam Earthquake of 1950. Braided Rivers, Special Publication No. 36, International Association of Sedimentologists.
Sarma, J. N. 2005. Fluvial process and morphology of the Brahmaputra River in Assam, India. Geomorphology, Elsevier, Vol. 70, issue 3-4, 226-256.
US Army Corps of Engineers – Hydrologic Engineering Center (USACOE-HEC). 2010. HEC-RAS River Analysis System, Version 4.1 CPD-69. US Army Corps of Engineers, Institute for Water Resources, Hydrologic Engineering Center, Davis, CA.
