Rational - Response of Flood Events to Land Use and Climate Change - page 7

methodology and assessments performed. Additionally, key recommendations are
discussed. Chapter 8 highlights some directions and perspectives for future inves-
tigation for the Barcelonnette area.
Hydrodynamic simulation, together with trend detection in climatologic and
hydrologic parameters and land use change estimation, led to the conception of a
modern structure for flood hazard assessment. This thesis aims to establish such a
system, using a case study catchment to test and demonstrate its ability to perform
under the typical conditions of mountainous catchments in France.
In this research, the focus was on flood hazard assessment based on hydrologic,
land use, and climatic parameter analysis in a time series. Within hydraulic
modelling, the output is affected to a large extent by model input parameters. In
this mention, first the testing of the quality of the DME using interpolation tech-
niques was important in order to accomplish a successful modeling. Then, all
geometric data were prepared in the HEC-GeoRAS model.
Before starting hydrodynamic modeling, based on available data, land use maps
from 1957 to 2008 were analyzed to gain better insight into river morphology
change over time and to access a variety of roughness derived from different land
use maps to assign to cross-sections in flood modeling. Then, the focus was on the
preparation of hazard maps and frequency analysis. The main purpose of this part
was to compare some severe flood events that occurred in this area, and then to
estimate flood propagation using hydrodynamic models (1D HEC-RAS model and
1D-2D SOBEK model), with respect to a variety of cross-sections, river mor-
phology, and different hydrographs. Moreover, the preparation of different hydro-
logic scenarios for civil protection purposes was considered in this study area: (1) In
the case of the DEM scenario, the model was run with different DEM resolutions to
understand how the grid resolution of a DEM affects the flood characteristics when
simulated by a 1D and 1D-2D flood models. (2) In the case of the river change
scenario, the aim was to evaluate river change scenarios in order to know how river
channel changes affect the flood, using different profiles and different discharge
values. (3) In the case of the dyke failure scenario, the major aspect was realizing the
possible effects of dyke failure. (4) Finally, the main aim in the model scenario, was
to compare two different hydrologic models, including HEC-RAS (1D) and SOBEK
(1D-2D), in order to know what the advantages (or strengths) and disadvantages (or
weaknesses) of these models are. In the last part of the hydrodynamic simulation, the
probability distribution and the return period were analyzed.
Climatic parameters are one of the most important triggering factors for a flood
event. The trend analysis of hydrologic and meteorologic series is important, and
even more relevant when considering the regional effects of global climate change.
Therefore, in the framework of climatic parameter analysis, any probable abrupt
points and trend analysis in the time series of climatologic and hydrologic
parameters using statistical methods were explored. Based on available aerial
photos, land use change was detected and statistically analyzed from 1957 to 2008
using GIS techniques. Finally, to gain feedback and understand the relationship
among climate change, land use change, and flood events, all achieved results were
reviewed. Therefore, the thesis is divided into three main phases: