Journal of African Earth Sciences 142 (2018) 44e51

Contents lists available at ScienceDirect

Journal of African Earth Sciences

journal homepage: [Link]/locate/jafrearsci

Spatiotemporal floodplain mapping and prediction using HEC-RAS -

GIS tools: Case of the Mejerda river, Tunisia
C. Ben Khalfallah a, S. Saidi a, b, *
a
Faculty of Sciences, University of Tunis El Manar, 2092, Manar II, Tunis, Tunisia
b
Water, Energy, Environment Laboratory ENIS, ENIS. B.P.“W”, 3038, Sfax, Tunisia

a r t i c l e i n f o a b s t r a c t

Article history: The floods have become a scourge in recent years (Floods of, 2003, 2006, 2009, 2011, and 2012),
Available online 6 March 2018 increasingly frequent and devastating. Tunisia does not escape flooding problems, the flood management
requires basically a better knowledge of the phenomenon (flood), and the use of predictive methods. In
Keywords: order to limit this risk, we became interested in hydrodynamics modeling of Medjerda basin.
HEC RAS To reach this aim, rainfall distribution is studied and mapped using GIS tools. In addition, flood and
GIS
return period estimation of rainfall are calculated using Hyfran. Also, Simulations of recent floods are
Modeling
calculated and mapped using HEC-RAS and HEC-GeoRAS for the most recent flood occurred in February
Prediction
Flood risk
eMarch 2015 in Medjerda basin.
Medjerda Basin Tunisia The analysis of the results shows a good correlation between simulated parameters and those
measured. There is a flood of the river exceeding 240 m3/s (DGRE, 2015) and more flowing sections are
observed in the future simulations; for return periods of 10yr, 20yr and 50yr.
© 2018 Elsevier Ltd. All rights reserved.

1. Introduction and floodplain sediment transport (Marriott, 1992; Pizzuto, 1987),

and river system hydrology (Dutta et al., 2013). Combined with
Numerical modeling of shallow water flow is an important hydrological models, and river models, the application of flood
research topic for flood risk management. The simulation results modeling has been extended to modeling that aims to formulate
play a significant role in national decision-making on flood pre- climate adaptation and risk mitigation strategies.
vention and control. One-dimensional shallow water models are Some studies which are carried out in Tunisia and have the aim
quite often preferred in the field of engineering (Fread, D.L., 1993; the protection of flooded areas using statistical methods, simula-
Ervine and MacLeod., 1999; HEC-RAS; ISIS; MIKE11), mostly due tions, hydrological models (Euchi, 2013; Selmi, 2013). On the other
to the fast computation (Syme, 2011) and its easy application (An hand, the innovation of this work when comparing to local studies
et al., 2015). Also 2D models are developed and used (i.e., Bates is the prediction of flooded areas for different return periods and
and De Roo, 2000; Yoon and Kang, 2004; Mignot et al., 2006; the mapping of flooded zones. So decision makers can intervene
George and LeVeque, 2008; Kim and Cho, 2011). effectively to protect areas with risk. and reduce the impact of
Worldwide Research studies since the 1970 have greatly floods in the future.
improved the capability of flood modeling. The models are widely Flood modeling has greatly improved in recent years with the
used in flood risk mapping and have been developed to simulate advent of geomatic tools and especially Geographic Information
flooded areas (Apel et al., 2006; Dutta et al., 2006; Baky et al., 2012), Systems (GIS). In fact, in this study the combination of HEC RAS,
flood damage assessment (Bhuiyan and Dutta, 2012; Merz et al., HECeGEORAS and GIS are used to delineate the flood inundation
2010), real-time flood forecasting (Arduino et al., 2005), and wa- zones, to proceed with a simulation for different dates.
ter resources planning (Vaze et al., 2013), as well as having served The specificity of this work is to shares data with the GIS, HEC
as an important prerequisite for investigating river bank erosion GEORAS and HEC RAS interfaces. Flood management should be
considered a spatial problem because flood intensities and char-
acteristics vary with geographic location (Foudi et al., 2015). As
* Corresponding author. Faculty of Sciences of Tunis, University of Tunis El Manar, newer GIS software was developed that improved geospatial
2092, Manar II, Tunis, Tunisia. visualization and computational efficiency, newer versions of HEC-
E-mail addresses: salwa_saidi@[Link], salwasaidifst@[Link] (S. Saidi).

[Link]
1464-343X/© 2018 Elsevier Ltd. All rights reserved.
 C. Ben Khalfallah, S. Saidi / Journal of African Earth Sciences 142 (2018) 44e51 45

GeoRAS followed (HEC, 2002; HEC, 2005; HEC, 2009; USACE, 2016).
So the main goal of this study is to estimate the extent of floodplain
inundations, corresponding to Sidi Salem outflows, in order to
create management plans for flood risk. Specific objectives of this
work are: i) Modeling and simulation of River reach extends from
the Sidi Salem dam of to Al Aroussia one using HEC RAS ii) flood risk
mapping using HEC GEORAS software iii) Estimation of flood zones
at return periods of interest.

2. Study area characteristics

2.1. Geographical location of the study area

Fig. 2. Comparison between rainfall measures in February 2015 and the mean Rainfall
in rainfall stations of the study area.
The Medjerda, the major Tunisian river (Ben Mansoura et al.,
2001), with its delta located on the western shore of the Gulf of
Tunis (Fig. 1). It originates in northeastern Algeria and then flows 2.2. Hydrologic characteristics of Medjerda River
eastwards to Tunisia then entering the Gulf of Utica of the
Mediterranean. with a catchment covers approximately For our study case, we opted for the middle valley of Medjerda.
24,000 km2, of which 7700 located in Algeria (Bouraoui et al., We will pay a particular attention to the reach located between the
2005), an approximate water flow of 30 m3 s
1, Water from two dams of Sidi Salem (the largest dam in North Africa) and Al
Medjerda is used for irrigation and is pivotal to the region's Aroussia (Gharbi and Soualimia, 2016) (Fig. 3). This reach is
agriculture (Mlayah et al., 2009), 90,000 ha are irrigated with extended over a length of 85 km; it is powered by tributaries which
water from the Medjerda to which must be added 18,000 ha are (Ben Khalfallah, 2015);
(DGRE, 2009). The climate is Mediterranean, characterized by
mild, rainy winters and hot, dry summers with no marked - The Siliana hydrographic network/Wadi: This is a major tribu-
intervening seasons. tary of the Medjerda which drains a basin with an area of
In fact, temperature is characterized by a maximum of 37  C to 2220 km2. The annual average supply to the dam is estimated at
38  C measured in July and August and a minimum of 2  Ce6  C in 6 50.106 m3 year
1 (Ben Khalfallah, 2015);
December and March.
The most important and recent event is the flood of February
2015 which will be considered in this study. In fact, it covers the
Medjerda Watershed especially Be
ja, Jendouba, Bousalem, Medjez
el Bab and Bizerte localities and caused two deaths. In fact, in
February the highest volume of rainfall is measured when
comparing to the rainfall mean (Fig. 2). The rainfall ranges between
70 mm and 180 mm in February 2015.
Consequently, these events proved that the risk of flood
downstream of Sidi Salem Dam remained very high. They also
emphasized the need for more reliable predictions of flood levels,
which in turn call for accurate simulation.

Fig. 1. Localization of the study area. Fig. 3. Hydrographic network tributaries of Medjerda River (Ben Khalfallah, 2015).
46 C. Ben Khalfallah, S. Saidi / Journal of African Earth Sciences 142 (2018) 44e51

Hydrology and Introducing discharge flood return

STEP 1 hydraulic canal rate and border period using
Hyfran

TIN
STEP 2 GIS HEC RAS
Data

STEP 3 Introducing Main Channel

Hydrography network geometric data

STEP 4 Modelling of Medjerda Flood Flows

Model

STEP 5 Considering output Preparing first report in order to flood level for
results several return
system (GIS) periods

STEP 6 GIS HEC Cartography

GeoRAS Ras

STEP 7 Flood Zoning Risk Mapping Flood Management

STEP 8 Results Suggests

Fig. 4. Flow chart showing the methodology adopted.

- The Khalled wadi: It starts in the Krib plain; it drains a catch- - The Zerga wadi: The watershed drained by this river covers an
ment area of 452 km2 to its confluence with the Medjerda area of 308 km2, the average annual inflows are estimated at
around Testour and in the upstream of the river in the conflu- 25.10 6 m3 year
1;
ence of Siliana Wadi with Medjerda River. The average annual - The Miliane wadi: it takes the name of Wadi el Kebir upstream.
contribution is 35,106 m3 year
1. Its confluence with Medjerda is It drains a catchment area of 2000 km2. It flows into the Gulf of
located in the immediate vicinity of Sidi Salem dam at an alti- Tunis between Rades and ez_Zahra. The length of its bed is
tude of 90 m; 150 km. Their right-bank tributaries are: Hamma Wadi, Wadi
- The Lahmar wadi: It drains the Goubellat plain; its watershed Rihane and Wadi Melah; left-bank tributaries are the Bou Arada
covers an area of approximately 520 km2. The contribution of and Haddada wadis. The average annual contribution of Wadi
this wadi is very low estimated at 6.106 year
1; Miliane is evaluated at 56.106 m3 year
1.
- The Oued Chafrou: It is located in the alluvial plain, starts and
flows into Medjerda at the City of Jedeida. The watershed, In order to examine changes in Medjerda River, 248 cross-
drained by this river, covers an area of approximately 592 km2 sections for channel survey are used in this work. These cross-
with an average rainfall of 400 mm year
1. sections were examined, in May 2008 (DGBGTH, 2007).
- Beja Wadi: This is a tributary of the left bank of Medjerda, its The observation of the cross sections and especially those of the
watershed is not very large and covers 206 km. The annual downstream show that river have been going considerable changes
average contributions are significant and estimated at 6 which are due to especially the silting of the Sidi Salem dam
56.106 m3 year
1; reservoir.
 C. Ben Khalfallah, S. Saidi / Journal of African Earth Sciences 142 (2018) 44e51 47

3. Materials and methods HEC-RAS, has been applied extensively in calculating the hy-
draulic characteristics of rivers (Carson, 2006; Pappenberger et al.,
3.1. Statistical calculation of return period of flood 2005). It is an integrated program used to calculate water surface
profiles using the following energy equation (Brunner, 2002):
To carry out the statistical tests and to calculate return period,
Hyfran software is used. HyfranPlus (HYdrological FRequency a1 V12 a2 V22
z1 þ y1 þ ¼ z 2 þ y2 þ þ he (1)
Analysis Plus Decision Support System) is a software used to fit 2g 2g
statistical distributions.
HYFRAN includes a number of powerful, flexible, user-friendly in which: у represents water depth (m); z represents channel
mathematical tools that can be used for the statistical analysis of elevation (m); V represents average velocity (m/s); a is a velocity
extreme events. HYFRAN can be used in any study that requires weighting coefficient; he represents energy head loss (m); g rep-
fitting of statistical distribution to an independent and identically resents gravitational acceleration (m/s2); and subscripts 1 and 2
distributed data series. Also, it includes a large number of proba- denote cross sections 1 and 2, respectively (Chihhao et al., 2009).
bility distributions such as; Type I Extreme value (Gumbel), General
Extreme value (GEV), Weibull, Normal, Log-normal, Pearson Type 3.3. Application
III and Log-Pearson Type III distributions.
In this case, it is used to estimate the future event (extreme The methodology applied in this study is the integration of GIS
events) of different return periods, to get the probability of the “HEC-GEORAS” and HEC-RAS flood simulation models (Fig. 4).
event for designing and to identity the event's statistical
characteristics.
3.3.1. Data acquisition
The return period usually represent the 2, 10, 25, and 50 year
Data for HEC RAS modeling is collected from different sources
storms. The return period is usually selected on the basis of level of
and in different forma (Table 1). All the data files have a common
protection.
spatial reference (UTM ZONE 32 Nord, WGS 1984).
The specificity of the modeling using HEC RAS is to integrate the
3.2. Modeling and simulation
shapefile data without problems which can permit to exploit and
update existing data.
3.2.1. Modeling of Medjerda flood flows
The choice of a specific model depends on the nature of the
problem and our objectif. According to the method of calculation, 3.3.2. Development of HEC-RAS flood simulation model
there are totally decoupled models, semi-coupled or fully coupled, The pre-processing of the geometric data and the post-
so we chose 1D model HEC RAS for processing and simulation of the processing of the outputs (to visualize the flooding impact) that
results. are required by the HEC-RAS model processes are done using HEC
Hydraulic modeling in Hec-RAS requires user input specifying GEORAS. This interface (an extension in Arc GIS software) allows
the channel geometry. Cross-sectional geometry including eleva- the preparation of geometric data import into HEC-RAS and pro-
tions, spacing and stream bank stations must be entered by the cesses simulation results exported from HEC-RAS in a geospatial
modeler and should be based on field measurements. environment. To achieve consistent accuracy through the process of
To initiate the collection of geometric input data, transverse integrating HEC-RAS and GIS, choosing the correct coordinate
cross-sections were staked and measured along the river's reach. system and mapping projection plays an important role. The TIN
So, in total 248 cross-sections were modeled. downloaded from SRTM has the North Africa Datum of 1984
(WGS84) with Transverse Mercator projection and these are
3.2.2. Description of the used model applied to the data frame for efficiency.
HEC-RAS is a widely used software application that perform one Cross sections used in HEC RAS simulations are created to
and two dimensional hydraulic calculations for a full network of extract elevation data from the TIN. These cross sections are crucial
natural and constructed channels, overbank/floodplain areas, pro- in HEC RAS simulations. It can provide the locations of bank sta-
tected areas; etc (Hydrologic Engineering Center, 2016). It has tions, down-stream reach lengths and manning values (land use
found wide application for analyzing flow in rivers and in flood codes).
plain studies.
HEC-RAS can offer four main modules (USACE, 2002): 3.3.3. Simulation of 2015 flood event in HEC-RAS
The processing of the flood simulation in HEC-RAS is done by
- steady flow water surface profiles, which is intended for calcu- using steady flow simulation. In this process, the boundary condi-
lating water surface profiles for steady gradually varied flow; tions are established at all the ends of the river nodes by entering
- Unsteady flow simulation; the normal depth value and initial conditions were set by adapting
- Sediment transport computations; - Water quality analysis; the model with the corresponding stream flows for 2015 flood
which is intended to allow the user to perform riverine water event. The HEC-RAS output is exported to HEC-Geo RAS for post-
quality analyses. processing of HEC RAS results.

Table 1
Data sources and mode of processing.

Step of modeling Data Source Forma Mode of processing

Modeling Topography SRTM (Shuttle Radar Topography Mission) Raster (img) Interpolation (TIN)
Hydrography network General Direction of Water Resources (DGRE, 2015) Raster (img) Digitalization
Cross section of river General Direction of Dam Tunisia (DGBGTH, 2015) Excel (xls) Import in HEC RAS
Calibration of the model Flow data General Direction of Water Resources (DGRE, 2015) Excel (xls) Import in HEC RAS
Prevision Flow data General Direction of Water Resources (DGRE, 2015) Excel (xls) Import in HEC RAS
48 C. Ben Khalfallah, S. Saidi / Journal of African Earth Sciences 142 (2018) 44e51

3.3.4. Post processing of HEC-RAS results complete geometric datasets of the river from TIN for the HEC-RAS
Post-Processing facilitates the automated floodplain delineation input, there are several rules and procedures in the HECGeoRAS/
based on the data contained in the HEC RAS- GIS output file and the HEC-RAS manual regarding how to digitize or create the river ge-
original TIN. Using HEC-Geo RAS functionalities, the imported HEC- ometry components. Allowed us the mapping of flood zone. In fact,
RAS results are processed with the TIN of the Medjerda to generate the digital elevations model (TIN) and the scheme of flood plain
the flood water surface extents and the flood water depth files for extension, are illustrated in the Fig. 7. The simulated water depth
2015 flood event. reaches more than two meters in some locations.
Disasters caused by flash floods usually occur in villages and
4. Results and discussion towns are situated along channels. The flood mapping indicated
that for low recurrence periods events some areas are highly sus-
4.1. The flood return period using Hyfran ceptible to flooding. A sharp increase in area flood extent in Medjez
el Bab and El Herri station was noted.
In this study Generalized Extreme Value (GEV) is used to esti- After dealing with the geometric data, flow data were imported
mate the flow during time intervals (return period). GEV distribu- into the HEC-RAS to simulate the model for the flood extensions for
tion is the limiting distribution for the minimum or maximum of a return periods of 20, 50 and 100 year return periods. With this
series of random variables. consideration, inundation simulation (extent and depth of flood)
The statistical adjustment used to estimate water flow is a long was calculated with 30 m resolution and mapped using HEC
observation database from 1981 to 2015, measured in Slouguia GEORAS.
station (Fig. 5). The result of adjustment of the rainfall database For the flow of 238 m3/s, the water is flow out in some sections
using the maximum likelihood method and the Generalized especially in Medjez El Beb, Andalous dam, El Herri and Borj
Extreme Value (GEV) is illustrated in the Fig. 5. Ettoumi localities. It can be explained by spontaneous habitats and
The estimated flows which will be used after that in simulations urban waste in the major bed of the wadi and the narrowing of the
are summarized in the Table 2. section and the reduction of the flow capacity.
After modeling and simulation steps, the data flow were im-
4.2. Simulation in HEC RAS ported in HEC-Geo RAS interface using a data exchange format
called a RAS GIS File in order to product the spatial variation of
After stimulating the flow in HEC-RAS Model, some character- floods as maps.
istics such as longitudinal profile can be observed and according to
them the organizing plans of river and flood control can be pro-
4.5. Prediction of future flood flows for different return periods
vided (Fig. 6a).
Fig. 6b presents an example of the cross section profiles with
Prediction is the main manner to manage food inundation and
water depths calculated with HEC-RAS.
to improve decision making in flood risk protection. RAS-GIS is
used to estimate flood return period for 5, 10, 20 and 50 yr.
4.3. Calibration and validation
The flow of flood calculated by means of Hyfran software for
different period is used to delineate the inundation zones.
Calibration and validation are done by using the hydrograph
It is reveals that flow increases through the time proportionally
during the flood seasons of 2015. Unsteady flow is the condition for
with the return period and consequently the inundation extent. In
simulating the floods. Flood plain area is defined by extending the
fact, in 2015 event the inundation zone is calculated as 11.68 Km2,
cross section. The flood is occurred when the river water is over-
for return period of 10yr the area of the inundation zone is of
flowing the flood plain. Triangulated irregular network (TIN)
18 Km2, for 20yr the inundation area is of 39.25 Km2, for return
premise is needed to perform the calculation of data. Peak levels
period of 50yr the inundation area increase and reach 49.7 Km2
and time of occurrence show a good quality of prediction. Mean
(Figs. 8e10).
square error is fall between 5 cm and 10 cm.
The flowing sections are concentrated especially in the city of
Medjez El Beb and at the Andalous Bridge.
4.4. Flood mapping The results indicate that climatic, physical and hydrological
characteristics are the major factors controlling frequent floods for
The ArcGIS extension of HEC-GeoRAS was used to extract the different return periods. In fact, the slope and the permeability
demonstrate a high influence on flood intensity and frequency. Also
land use and its variation through time can influence infiltration
capacity and flow intensity. It is generally accepted that changes in
land-use (Urbanization because towns and cities have more
impermeable surfaces, deforestation, because removing trees re-
duces the amount of water intercepted and increases run-off, all
have a part to play. etc.) (Sarhadi et al., 2012; Wahren et al., 2009).
Also another important factor can interfere which is the climatic
one. For different return periods when comparing flooded area
with projections of temperature we can note a similarity in evo-
lution. So, we can conclude that the global exposure to floods will

Table 2
Annual flow estimated in function of each return period (T).

Return period (year) State 2015 5 10 20 50

Mean annual flow estimated m3 s
1 240 300 450 600 800
Fig. 5. The result of non-exceedance probability using monthly rainfall data.
 C. Ben Khalfallah, S. Saidi / Journal of African Earth Sciences 142 (2018) 44e51 49

a)

b)

Fig. 6. Simulation in HEC RAS; a) Example of HEC-RAS calculated water depths and cross section profiles of Mejerda River; b)View 3D Multiple Cross Section Plot.

increase depending on the degree of warming. Furthermore, the effects of resolutions of elevation datasets and
Manning's roughness in prediction of travel time and flood inun-
5. Conclusions dation areas were investigated. The study utilized a broadly
accepted tool, HEC-RAS to perform the required hydraulic simula-
Obtaining flood information in data-poor regions is a difficult tion. A HEC-GeoRAS, an ArcGIS extension, was used to produce
and uncertain task, but is very valuable for flood risk management flood inundations maps.
in order to protect human civilization and the environment. In this The one-dimensional hydraulic model permitted to simulate the
study, an approach for flood warning system was introduced flow events that occurred at the Medjerda River. The model was
including digital map preparation and travel time computation. used to calculate the respective water profiles for the major peak
50 C. Ben Khalfallah, S. Saidi / Journal of African Earth Sciences 142 (2018) 44e51

Fig. 7. Depth grid of the flood waters simulation using Hec GEORAS (state of February Fig. 9. Inundation extent and flood water depth in Medjerda river (Sidi Salem dam- El
2015 event). Aroussia Dam) for return period of 20 years.

Fig. 8. Inundation extent and flood water depth in Medjerda river (Sidi Salem dam- El
Fig. 10. Inundation extent and flood water depth in Medjerda river (Sidi Salem dam- El
Aroussia Dam) for return period for 10 years.
Aroussia Dam) for return period of 50 years.

flows recorded during the floods having occurred between Febru- of a large dam, because of the meteorological and hydrological
aryeMarch 2015, also water profiles were calculated to estimate extreme phenomena.
periods of flood return. The ArcGIS extension of HEC-GeoRAS The hydraulic model was setup for various return period floods
permitted establishing the extension flood maps simulated for using different elevation datasets and a range of possible roughness
each calculated water profile. The analysis of the results showed values to observe the uncertainties involved in flood inundation
that the preparatory management for this season has significantly mapping process. A topographic survey was conducted to get ac-
attenuated the extent of flooding during this period. The use of this curate elevation dataset in the river channel sections, this meth-
kind of results based on hydraulic simulation and geographic in- odology can help to better understand the various levels of
formation system can contribute meaningfully to develop strate- protections needed to minimize and prevent damage from river
gies to fight against flood hazards that persist despite the existence systems; thus, the simulation of the river hydraulics and the
 C. Ben Khalfallah, S. Saidi / Journal of African Earth Sciences 142 (2018) 44e51 51

visualization of the flood inundation in probable scenarios can be d'inondations sur l’oued [Link] report, p. 90.
DGRE (General Directorate for water Resources), 2009. Etude des risques d'inon-
useful. This technology advances our ability to model the hydraulic
dations sur l’oued Medjerda en [Link] report, p. 180.
processes in frequent and infrequent (dam-break) flood events. Dutta, D., Herath, S., Musiake, K., 2006. An application of a flood risk analysis system
Therefore, these data sources are far more accurate and in fact quite for impact analysis of a flood control plan in a river basin. Hydrol. Process. 20
useful in spatially detailed floodplain inundation studies. The pre- (6), 1365e1384.
Dutta, D., Teng, J., Vaze, J., Lerat, J., Marvanek, S., 2013. Storage-based approaches to
sent model will have the capability to perform hydrological studies build floodplain inundation modelling capability in river system models for
on a regional scale and can be incorporated into or provide water resources planning and accounting. J. Hydrol 504 (0), 12e28.
boundary conditions for local models as well. Ervine, D.A., MacLeod, A.B., 1999. Modelling a river channel with distant floodbanks.
Proc. Inst. Civil Eng. Water Maritime Energy 136, 21e33.
Inundation maps are useful for preparedness before the occur- Foudi, S., Ose
s-Eraso, N., Tamayo, l., 2015. Integrated spatial flood risk assessment:
rence of floods, timely response to future floods, damage assess- the case of Zaragoza. Land Use Pol. 42, 278e292.
ment, mitigation and flood risk analysis. These tools act as an Fread, D.L., Hsu, K.S., 1993. Applicability of two simplified flood routing methods:
level-pool and Muskingum-Cunge. In: ASCE National Hydraulic Engineering
important guideline for decision makers, especially when coupled Conference, pp. 1564e1568. San Francisco, 26-30 July 1993.
with skills, experience and proper judgments, to plan accordingly George, D.L., LeVeque, R.J., 2008. High resolution methods and adaptive refinement
for future probable flood disasters. for tsunami propagation and inundation. In: Hyperbolic Problems: Theory,
Numerics, Applications. Springer, Berlin, Germany, pp. 541e550. Proc. 11th
The study shows that the HEC-RAS is an important tool for [Link]. On Hyperbolic Problems, Lyon, France, 17-21 July 2006.
studying and understanding flood events. Future applications of Gharbi, M., Soualimia, A., 2016. Floods effects on rivers morphological changes
the new HEC-RAS version 5 may help to analyze possible flood application to the Medjerda River in Tunisia. J. Hydromech. 64, 56e66. https://
[Link]/10.15115/johh-2016-0004, 2016, 1.
management strategies.
HEC, 2002. HEC-GeoRAS e an Extension for Support of HEC-RAS Using ArcView,
Also other climatic factors can be added in further studies to CPD-76, October 2002. Hydrologic Engineering Center, Institute for Water Re-
highlight the impact of global changes on flooding risk. sources, U.S. Corps of Engineers, Davis, CA.
HEC, 2005. HEC-GeoRAS e an Extension for Support of HEC-RAS Using ArcGIS (v8.3/
9.1), CPD-83, September 2005. Hydrologic Engineering Center, Institute for
Acknowledgements Water Resources, U.S. Corps of Engineers, Davis, CA.
HEC, 2009. HEC-GeoRAS e an Extension for Support of HEC-RAS Using ArcGIS (v9.2/
The author thanks Mr. Mondher BEN RJAB of the General 9.3), CPD-83, September 2009. Hydrologic Engineering Center, Institute for
Water Resources, U.S. Corps of Engineers, Davis, CA.
Directorate of Water Resources of the Ministry of Agriculture for the HEC, 2016. HEC-RAS River Analysis System, User's Manual, Version 5.0, CPD-68,
collaboration and the analysis of data and model results. February 2016. Hydrologic Engineering Center, Institute for Water Resources,
U.S. Corps of Engineers, Davis, CA.
Euchi, K. , 2013. Protection contre les inondations de la ville de Kasserine. Institut
References National Agronomique De Tunisie, p. 195.
Kim, H., Cho, Y., 2011. Numerical model for flood routing with a Cartesian cut-cell
An, H., Yu, S., Lee, G., Kim, Y., 2015. Analysis of an open source quadtree grid shallow domain. J. Hydraul. Res. 49 (2), 205e212.
water flow solver for flood simulation. Quat. Int. 384, 118e128. [Link] Marriott, S., 1992. Textural analysis and modeling of a flood deposit - river Severn.
10.1016/[Link].2015.01.032. UK. Earth Surf. Process. Landf. 17 (7), 687e697.
Apel, H., Thieken, A., Merz, B., Blo €schol, G., 2006. A probabilistic modelling system Merz, B., Kreibich, H., Schwarza, R., Thieken, A., 2010. Review article ‘Assessment of
for assessing flood risks. Nat. Hazards 38 (1e2), 79e100. economic flooddamage’. Nat. Hazards Earth Syst. Sci. 10 (8), 1697e1724.
Arduino, G., Reggiani, P., Todini, E., 2005. Recent advances in flood forecasting and Mignot, E., Paquier, A., Haider, S., 2006. Modeling floods in a dense urban area using
flood risk assessment. Hydrol. Earth Syst. Sci. 9 (4), 280e284. 2D shallow water equations. J. Hydrol. 327, 186e199.
Baky, A.A., Zaman, A.M., Khan, A.U., 2012. Managing Flood Flows for Crop Produc- Mlayah, A., Ferreira da Silva, E., Rocha, F., Ben Hamza, Ch, Charef, A., Noronha, F.,
tion Risk Management with Hydraulic and GIS Modeling: Case Study of Agri- 2009. The Oued Melle gue: mining activity, stream sediments and dispersion of
cultural Areas in Shariatpur APCBEE Procedia, pp. 318e324. base metals in natural environments, North-western Tunisia. J. Geochem.
Bates, P.D., De Roo, A.P.J., 2000. A simple raster-based model for flood inundation Explor. 102, 27e36.
simulation. J. Hydrol. 236 (1e2), 54e77. Pappenberger, F., Beven, K., Horrit, M., Blazkova, S., 2005. Uncertainty in the cali-
Ben Mansoura, A., Garchi, S., Daly, H., 2001. Analyzing forest users' destructive bration of effective roughness parameters in HEC-RAS using inundation and
behavior in Northern Tunisia. Land Use Pol. 18 (2), 153e163. [Link] downstream level observation. J. Hydrol. 302, 46e69.
10.1016/S0264-8377(01)00004-7. Pizzuto, J.E., 1987. Sediment diffusion during overbank flows. Sedimentology 34 (2),
Ben Khalfallah, C., 2015. Contribution de la mode
lisation hydraulique et des SIG a
 la 301e317.
cartographie et la pre
vision des zones inondables : Cas de la basse valle
e de la Sarhadi, A., Soltani, S., Modarres, R., 2012. Probabilistic flood inundation mapping of
Medjerda. Me
moire de Maste re. Faculte

des Sciences de Tunis, Universite
Tunis ungauged rivers: Linking GIS techniques and frequency analysis. J. Hydrol. 458-
El Manar, p. 155. 459, 68e86.
Bhuiyan, M., Dutta, D., 2012. Analysis of flood vulnerability and assessment of the Selmi, M., 2013. Protection contre les inondations des Zones Nord et Est du Grand
impacts in coastal zones of Bangladesh due to potential sea-level rise. Nat. Tunis. Institut National Agronomique De Tunisie, p. 90.
Hazards 61 (2), 729e743. Syme, B., 2011. Pros and cons of 1D and 2D modeling. In: Floodplain Management
Bouraoui, F., Benabdallah, S., Jrad, A., Bidoglio, G., 2005. Application of the SWAT Association Conference. Sheraton San Diego Hotel & Marina, San Diego, USA.
model in the Medjerda river basin (Tunisia). Phys. Chem. Earth 30, 497e507. US Army Corps of Engineering (USACE), 2002. HEC-RAS. River Analysis System.
Brunner, G.W., 2002. HEC-RAS, River Analysis System Hydraulic Reference Manual. Hydraulic Reference Manual Version 3.1.
Hydrological Engineering Center, US Army Corps of Engineers, Davis, CA. US Army Corps of Engineers, Davis California, 2016. Hec-RAS. River Analysis System.
Carson, E.C., 2006. Hydrologic modeling of flood conveyance and impacts of historic Hydraulic Reference [Link] 5.0.
overbank sedimentation on West Fork Black's Fork, Uinta Mountains, north- Vaze, J., Viney, N., Stenson, M., Renzullo, L., Van Dijk, A., Dutta, D., Crosbie, R.,
eastern Utah, USA. Geomorphology 75, 368e383. [Link] Lerat, J., Penton, D., Vleeshouwer, J., Peeters, L., Teng, J., Kim, S., Hughers, J.,
[Link].2005.07.022. Dawes, W., Zhang, Y., Leighton, B., Perraud, J.-M., Joehnk, K., Yang, A., Wang, B.,
Chihhao, F., Chun-Han, K., Wei-Shen, W., 2009. An innovative modeling approach Frost, A., Elmahdi, A., Smith, A., Daamen, C., 2013. The Australian water
using Qual2k and HEC-RAS integration to assess the impact of tidal effect on ressource assessment system (AWRA). In: Proceedings of the 20th International
River Water quality simulation. J. Environ. Manag. 90, 1824e1832. https:// Congress on Modelling and Simulation (MODSIM2013). Adelaide, Australia.
[Link]/10.1016/[Link].2008.11.011. Wahren, A., Feger, K.-H., Schwarzel, K., Munch, A., 2009. Land-use effects on flood
DGBGTH (General Directorate of dams and major Hydraulic works), 2015. Risque generation e considering soil hydraulic measurements in modeling. Adv.
d'Inondation sur l'Oued Medjerda, inedit report, p. 20. Geosci. 21, 99e107.
DGBGTH (General Directorate of dams and major Hydraulic works), 2007. Annu- Yoon, T.H., Kang, S.K., 2004. Finite volume model for two-dimensional shallow
aires des profils en travers en long de l'oued Medjerda, Official numerical file. water flows on unstructured grids. J. Hydraul. Eng. 130 (7), 678e688.
DGRE (General Directorate for water Resources), 2015. Etude des risques