الفهرس 
INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 130 |  |  |
| | | from | 130 | | |
Abstract
This research develops a general technique for performing flood plain and flood risk analysis
in floodways. The general technique consists of two steps. The first step involves the hydraulic
modeling of the floodway under consideration while the second step is the flood risk assessment.
The hydraulic modeling (i.e., the first step) is done via the application of a hydraulic model. The
chosen hydraulic model can vary from a 1D hydraulic model to a fully 2D hydrodynamic model
depending on the complexity and the extent of river geometry. For that reason, the thesis applies
different hydraulic models HEC-RAS-1D, FESWMS, and HEC-RAS-2D and provides a technical
reference for setting up and running these models. HEC-RAS-1D represents the family of onedimensional
models, FESWMS represents the group of fully 2D hydrodynamic models, while
HEC-RAS-2D represents the Semi-hydrodynamic 2D models. The thesis also provides a
comparative analysis between the three models in terms of accuracy and engineering applicability.
The results of the comparative analysis show that the choice of the hydraulic model must suit the
case study under consideration. Each model has its advantages, disadvantages, and best applicable
case. The 1D and semi-2D models proved to best suit the cases of long rivers (especially those
with steep slopes) due to their numerical stability in performing the hydraulic simulation. The fully
hydrodynamic group crashes and fails to model such long steep rivers due to the extreme
difficulties to spin down the water surface elevation in such steep and long cases. However, the
fully hydrodynamic group best simulates short reaches with complicated two-dimensional
configurations (e.g., flow in the vicinity of bends or bridges) due to their high accuracy in
representing the case without the need to spin down the model. The results also show that the
semi-2D models outperform the 1D-models in representing long rivers due to the less effort
encountered in representing river geometry and the more accuracy in representing the lateral flow
components. The thesis finally presents a novel GIS-based methodology to assess the flooding
risk. The methodology depends on using the specific energy of the flow (rather than the water
depth or velocity alone) to estimate the Flood Risk Index Factor (FRI) taking also vulnerability
into consideration. The analysis performed into this thesis is presented on a case study from the
Nyamwambe River in Uganda to illustrate the findings.