River Flood Modelling of Himayalan Rivers

Mid-semester Report

By

Ravindra Pratap Singh

(200104089)

Prabhav Maheshwari
(200104077)

Underthe guidance of

Prof. Subashisa Dutta

O
of

Tothe
DEPARTMENT OF CIVIL ENGINEERING
INDIAN INSTITUTE OF TECHNOLOGY GUWAHATI
GUWAHATI - 781039, ASSAM
Introduction:
Flood river modeling is a critical component of modern water resource management and disaster
mitigation strategies. Rivers, while essential for sustaining ecosystems and human societies, can also
pose significant threats when they overflow their banks, causing widespread flooding and
devastation. As climate change and urbanization alter precipitation patterns and land use, the
potential for more frequent and severe river floods becomes a growing concern. To effectively plan
and implement flood control measures, it is imperative to comprehensively understand and model
the complex dynamics of river systems.

Literature review
In recent years, river floods have become increasingly destructive due to rapid development, climate
change, and anthropogenic activities. To address these challenges, advanced modeling techniques
have played a pivotal role in flood management. In the Himalayan foreland basin, machine learning
algorithms like Random Forest, Support Vector Machine, and Extreme Gradient Boosting have been
deployed for flood susceptibility analysis, aiding in the identification of high-risk areas [1]. Similarly,
on the southern slope of the Himalaya, hazard mapping and risk assessments using the Hydrologic
Engineering Center's River Analysis System (HEC-RAS) have improved decision-making for water
resource management and flood control [2]. Numerical modeling techniques have also been
instrumental in tropical regions, offering insights into the causes and effects of floods, with 3D
modeling emerging as a solution to enhance vertical accuracy in floodplain analysis [3]. Moreover,
climate change's influence on floods has been scrutinized in the Brahmaputra basin using CMlP5
decadal predictions, revealing potential increases in flood intensity and frequency due to changes in
precipitation patterns [4]. This comprehensive literature review underscores the importance of
advanced modeling and climate science integration for effective flood risk assessment and
management, emphasizing the need for multidisciplinary approaches to tackle the complex
challenges posed by evolving flood behavior in the face of climate change.

The primary objectives of river flood modeling:

Predict Water Level Dynamics: Develop a reliable model to predict the water level dynamics of a
river system in response to varying discharges, channel geometry, and meteorological conditions.
This prediction will be a fundamental component of flood forecasting and early warning systems.
Assess Flow Velocity and Celerity: Calculate and analyse flow velocity, celerity (wave propagation
speed), and other hydraulic parameters to understand how water moves through the river network.
This information is crucial for assessing flood propagation and inundation patterns.

Analyse Channel Morphology: Investigate how the river channel morphology evolves under different
flow conditions, including the potential for erosion, sediment deposition, and changes in channel
cross-sections. This analysis can help in assessing the long-term impact of floods on the river's
physical characteristics.

Expectations from a River Flood Modeling Analysis:

Accurate Water Level Prediction: The analysis should provide precise predictions of water level
variations in the river system, considering factors such as river discharge, channel geometry, and
meteorological conditions. These predictions should be essential for reliable flood forecasting and
early warning systems.
Detailed Flow Velocity Assessment: The model should offer a comprehensive assessment of flow
velocities throughout the river network. Understandingthe velocity patterns is critical for assessing
flood propagation, potential hazards, and areas of high flood risk.

Comprehensive Celerity Analysis: Analyze celerity, or wave propagationspeed, within the river
system to determine how quickly flood waves travel downstream.This informationis vital for
estimating flood arrival times at various locations, aiding in timely evacuation and emergency
response planning.

Channel Morphology Insights: Investigate the impact of different flow conditions on river channel
morphology. Analyze erosion, sediment deposition, and changes in channel cross-sections to assess
the river's long-term response to flooding events. This insight can inform strategies for managing and
mitigating riverine erosion.
Sensitivity Analysis: Conduct sensitivity analyses to understandthe model's response to variations in
input parameters, such as precipitation patterns and discharge rates. Identify which parameters have
the most significant influence on flood predictions and channel behavior.

Calibration and Validation: Validate the model's accuracy by comparing its predictions with historical
flood data, if available. Perform calibration to refine the model's parameters, ensuring that it reliably
simulates real-world conditions.
Scenario Testing: Test the model under various flood scenarios, including different discharge rates,
storm intensities, and river geometries. Assess how these scenarios impact water levels, flow
velocities, and channel morphology to prepare for a range of potential flood events.

Visualization and Reporting: Develop effectivevisualizationtools and reportsto communicate

modeling results to stakeholders, including emergency response teams, local authorities, and
affected communities. These should facilitate informed decision-making and flood preparedness
efforts.
Historical Approaches to River Flood Modeling:
In the past, river flood modeling primarily relied on simplistic procedures that did not fully account
for the complex dynamics of flooding events. These traditional approaches, while providing some
insights, had several limitations:

Hydrological Simplification: Traditional river flood modeling often simplified the hydrological
processes, representing river flow and precipitation as static or simplified time-averaged inputs. This
approach did not adequately capture the temporal and spatial variability of rainfall and river flow.

Static Representation of Flooding: River flood models often treated floodwaters as static forces
without considering the dynamic nature of flood propagation. It often overlooked the complexities of
flood routing and wave propagation within river networks.

Simplistic Channel Morphology: Traditional models sometimes oversimplify river channel

geometries and morphological changes, neglecting the dynamic interactions between the riverbed
and water flow. This oversimplification limited the accuracy of flood extent predictions.

Working of Model:

of newt (c'

reue..za•.e

Of
Rots ve«cctv
and"-net •cccrurgto

The described MATLAB model is a valuable tool for the comprehensive analysis of water levels and
flow velocities within hydraulic systems. Beginning by importing essential data from an Excel sheet,
such as initial water levels and flow rates, the model conducts a thorough analysis, incorporating
values and z-values from the spreadsheet. This analysis yields critical insights into hydraulic
properties.

The model's iterative process calculates various section properties, such as cross-sectional area,
perimeter, conveyance, slope, and velocity, based on initial assumptions for water levels and
the
hydraulic head (H). It employs the secant method, as advised in the HEC-RAS manual, to refine
the accuracy of
calculations. Throughout the process, termination conditions are checked to ensure
enhancing our
results. Ultimately, the model provides a detailed profile of water behavior,
resource
understanding of hydraulic systems and aiding in critical decisions regarding water
management and flood control.
Preliminary results:

Water00th Cross Section

02

Cens

Observations and Limitations:

In practical flood events, the limitations of traditional river flood modeling became apparent, with
discrepancies between predicted and observed flood behaviors.

These models often struggled to account for the rapid onset and intensification of flash floods, which
are prevalent in many regions.

The omission of detailed topographic and hydraulic data, as well as the lack of consideration for local
geomorphological features, hindered the precision of flood inundation predictions.

Images:
Study area representing Brahmaputra basin for flood inundation assessment (Base map from
OpenStreetMap and OpenStreetMap Foundation).
https:/ldoi.org/_10.1371 [journal.pone.0273384.g001

References:
[IJ : Supriya Ghosh, Soumik Saha, Biswajit Bera, (2022), Flood susceptibility zonation using
advanced ensemble machine learning models within Himalayan foreland basin,
[2]: Aryal, Dibit & Wang, Lei & Adhikari, Tirtha & Zhou, Jing & Li, Xiuping & Shrestha, Maheswor &
Wang, Yuanwei & Chen, Deliang. (2020). A Model-Based Flood Hazard Mapping on the Southern
Slope of Himalaya.

[3]: Mohd Talha Anees, K. Abdullah, M.N.M. Nawawi, Nik Norulaini Nik Ab Rahman, Abd. Rahni Mt.
Piah, Nor Azazi Zakaria, M.I. Syakir, A.K. Mohd. Omar, (2016), Numerical modeling techniques for
flood analysis

[4]: Tushar Apurv, Rajeshwar Mehrotra, Ashish Sharma, Manish Kumar Goyal, Subashisa Dutta,
(2015) Impact of climate change on floods in the Brahmaputra basirsing CMlP5 decadal predictions