Life Science Journal, 2012;9(1) http://www.lifesciencesite.

com

The Hydrology of Wadi Ibrahim Catchment in Makkah City, the Kingdom of Saudi Arabia: The Interplay of
Urban Development and Flash Flood Hazards

Mohammed El bastawesy1, Khaled El Harby1,2, Turki Habeebullah2,3

1
Geography Department, Umm Al-Qura University, Makkah, Saudi Arabia
2
Centre of Research and Excellence in Hajj and Omrah (HajjCore), Umm Al-Qura University, Makkah, Saudi
Arabia.
3
The Custodian of the Two Holy Mosques Institute for Hajj Research, Umm Al-Qura University, Makkah, KSA
[email protected]; [email protected]

Abstract: This paper investigates the development of a hydrological model for the dryland catchment of Wadi
Ibrahim, which hosts the Holy Mosque of Makkah, in order to assess the interaction of urbanizing the alluvial
channels and the flow discharges of occasional flash floods. The original landscape and landcover of the catchment
have witnessed significant changes during the past few decades, where most of the alluvial channels and the
mountain footslopes have been covered by urban. The infrequent threat of flash floods has prompted the
development of a mitigation measure including; installation of rainfall-sewage system and subsurface culverts, in
addition to the conveyance of flows from the upper sub-catchment into another drainage basin. However, the latest
flash flood event of 30th of December 2010 has resulted in fatalities and demonstrated the insufficiency of the
current mitigation system to control flash floods. The runoff coefficient was estimated from the opportunistic
observations and measurements of the flow discharge parameters for the latest event, in addition to the recorded
rainfall parameters. The digital elevation model (DEM) was analyzed using Geographic Information System (GIS)
to determine the spatially distributed time-areas zones of the catchment, which were used to simulate the runoff
hydrographs under certain runoff coefficients and designed storms of long return periods. The development of
urban areas on expense of the alluvial channels resulted in a significant surge of runoff discharge, and therefore
increasing the threat of flash floods on urban areas downstream. As a result the transmission loss is diminishing;
thus raising the alarm on the potential recharge to the underlying alluvial aquifer of the sacred well of Zamzam.
Therefore, it is suggested that several small dams to be constructed at the fingertip drainage channels; to retain
considerable amount of water and sediment within the catchment and to act as point-source recharge to the alluvial
aquifer.
[Mohammed El bastawesy, Khaled El Harby and Turki Habeebullah. The Hydrology of Wadi Ibrahim Catchment
in Makkah City, the Kingdom of Saudi Arabia: The Interplay of Urban Development and Flash Flood
Hazards. Life Science Journal 2012; 9(1):580-589]. (ISSN: 1097-8135). http://www.lifesciencesite.com.86

Key Words: flash floods, Makkah, Kabaa, Wadi, Remote Sensing, GIS, Hydrology, Urban, Alluvial channels.

1. Introduction: surfaces; building, roads, storm sewers and paving

The drylands are characterized by droughts and usually decreases the infiltration capacities of the
the scarcity of water resources, occasional heavy underlying soils, and it significantly increase runoff-
storms often develop severe flash floods that discharge downstream. The estimated increase of
devastate the inhabited areas (Cooke et al., 1982; runoff coefficients; higher flow peaks and the
Tooth, 2000; Foody et al., 2004). The records of decrease of time to peak in urban catchment Depends
destructive flash floods and the notable regional on the extent of urbanization and the anthropogenic
frequencies have largely raised the public awareness managements of runoff of these catchments ( Suriya
and motivated more scientific research on the and Mudgal, in press (2011). In completely
hydrological processes within the fluvial system (e.g. impervious and fully sewered areas, peak discharge
Walling and Gregory, 1970; Chin and Gregory, increases 6 times more than in non-urbanized areas
2001). Originally, flash flood frequencies and (leveson, 1980), and 90 % of the total rainfall may be
magnitudes are controlled by the interplay of converted into urban runoff (Shang and Wilson,
different natural variables, including precipitation, 2009). However, the assessment of urbanization
antecedent conditions of the catchment, distribution impact on the hydrological response of the developed
of alluvium and water storage areas, etc. However, catchments is complex and hampered by the lack of
the growth of urbanization within the catchments can hydrological parameters and measurements, and the
be added as an additional significant controlling non-systematic temporal changes of the landuse and
factor for the development of flash flood. landcover due to urbanization (Smith and Bedient,
Urbanization, through the construction of impervious 1981).

580
 Life Science Journal, 2012;9(1) http://www.lifesciencesite.com

Generally, the hydrological data collected in parameters from the DEM has gradually replaced the
drylands remain insufficient and limited due to traditional manual delineation of these parameters
technical, political and economical factors, and from the conventional topographic maps (Band,
therefore, the hydrological processes are not fully 1986; Chorowicz et al., 1992). The manual method is
understood and the hydrological models are a tedious and error-prone technique in dryland
uncalibrated (El Bastawesy et al., 2009). alluvial areas, where the thalweg or active
Nonetheless, most of these data are also limited and longitudinal channel courses are not marked on
gathered by individuals or entities during pilot topographic maps and significant changes in these
projects and various independent case studies, which courses can take place over relatively short periods of
cannot represent the diversity within dryland setting time. The major issues associated with the derivation
and processes. It is also difficult to monitor flash of surface drainage networks from DEM are related
floods that occur suddenly, and the potential to to the quality, source and resolution of the DEM and
capture during classical-field experiments is very to the processing techniques and algorithms
rare. Therefore, the investigation of most flash employed (Zhang and Montgomery, 1994; Wolock
flood-events was mainly based on post-event and Price, 1994). However, the delineation of various
survey, which plays a critical role in gathering morphometric parameters (for a typical dryland
essential observations and data (Borga et al., 2008). catchment) was not very sensitive to the change of
The spatial and temporal variability of flash flood DEM resolution (from 20 m to 90 m) (El bastawesy,
events even within a single catchment clearly 2007).
demonstrate the need for the development of The aim of this paper is to develop a
distributed model; to assess vulnerability and to hydrological model for a typical dryland catchment,
mitigate against future damage. which has undergone dramatic urban expansion, in
Remote sensing data and Geographic order to assess the impact of flash flood hazard on
Information System (GIS) techniques are widely used urban areas, and also to determine the effect of
to estimate various distributed hydrological constructed mitigation measures on the fragile
parameters for the investigation of catchment surface and groundwater resources of the area. Herein
hydrology following given techniques and the Wadi Ibrahim catchment of Makkah city in the
interpretations (Schultz, 1987; Scipal et al., 2005; Kingdom of Saudi Arabia is selected for this study,
Milzow et al., 2008 ). The most straightforward use due to its importance as it hosts the Holy Mosque of
of remote sensing images is to identify geologic, Makkah and considerable in situ data are available.
geomorphologic and landuse-land cover features,
which in turn have a strong influence on overland The study area:
flow generation. Several remote sensing- products Makkah city is located in the southwestern part
such as the Tropical Rainfall Monitoring Mission of Al Hijaz province of the Kingdom of Saudi
(TRMM) and the Global Precipitation Climatology Arabia, between the low-lying coastal plain (Tihamat
Project (GPCP) are increasingly available to measure Al Hijaz) and the escarpment of the rugged Sarawat
the precipitation at semi-global coverage with a grid mountains that has resulted from the uplift associated
spatial resolution of 0.250 X 0.250 (Huffman et al., with the Red Sea rifting. The mountains ranges of
2007; Hossain et al., 2011). The remote sensing- Makkah are structurally controlled; they are aligned
rainfall estimates are widely utilized in the in northwest-southeast (e.g. Mena Mountains) or
hydrological models, particularly over the areas with east-west directions (e.g. El Tarqi Mountains).
poor or no rain-gauges data (e.g. Milewiski et al., However, few mountains are isolated and conform
2009; Abu El Magd et al., 2010). However, the semi-circular shape such as Thour Mountain (755 m)
uncertainty of remote sensing- rainfall estimates on and El Nour Mountain (642 m). The chronology of
flood prediction has to be considered; the calibration underlying rocks units in Makkah area is complex;
with in situ data in different gauged areas showed the isolated outcrops of amphibolites, gneiss and
non-systematic overestimation and underestimation schist are of uncertain stratigraphic relationship to
(Almazroui, 2011). Although, the remote sensing- other layered rock units of the mountain ranges, and
based rainfall estimates may represents the sole thus they are left unassigned on the geological map of
source of precipitation input to any hydrological the area (Fig 1). The predominant rock mass
model for the dryland catchments. underlying the catchment area of Wadi Ibrahim
The digital elevation models (DEM) are mainly belongs to the Precambrian, and mainly composed of
obtained from different sources and at different quartz diorite and tonalite (Kamil Suite). These
spatial resolution, and therefore are widely being outcrops are of moderate to steep relief, and also
used in the various hydrological models. The range from massive to well deformed and foliated
automatic delineation of catchment-hydrographic plutons. The Wadis of Makkah dissecting the

581
 Life Science Journal, 2012;9(1) http://www.lifesciencesite.com

mountain ranges are characterized by complex and fractured and weathered basement rocks (i.e. the
interlocking patterns. This complex pattern of lower half). The well is heavily pumped to provide
intersecting alluvial areas are very common in the millions of visitors with sacred water which is
dryland setting and reflect the morphotectonic available throughout the Holy Mosque via water
evolution of these drainage basins, where the paleo- fountains and dispensing containers.
channels were used to flow through different The earliest available record of rainfall for
directions than the contemporaneous flow pathways Makkah area is dated back to 1966, when the Saudi
(e.g. El Bastawesy et al., 2010). The Quaternary wadi Meteorological Authority installed a gauging station
alluvium consists of unconsolidated, moderately to at Umm Al-Gud of Makkah. However another
poorly sorted sand and gravel. The alluvium of the rainfall gauging station was installed by the
upper reaches of wadi courses is relatively thin and University of Umm Al-Qura at the main campus in
very poorly sorted, but it becomes well sorted and 1989, the rainfall data remains limited. Therefore,
reaches considerable thickness in the lower reaches. The Custodian of the Two Holy Mosques has
installed 2 additional automatic network weather
stations around Makkah since 2000 to deliver deliver
accurate, and reliable meteorological measurements
that can be used in hydrological applications. Overall,
the available rainfall data of Makkah area shows a
very high spatial and temporal variability. For
example the total annual rainfall in 1980 was less
than 5 mm, while the total rainfall of 1969 was 318
mm of which more 269 mm was precipitated in one
single storm (Fig 2). Although surge in local rainfall
magnitudes over the catchments of Makkah city is
vital to the recharge of the underlying aquifer, but it
usually develops severe and unfavorable flash floods.
For example, a destructive flash flood has occurred
in 1969 following the precipitation of more than 269
mm of rainfall over Makkah; the water partially filled
the holy mosque and stood 2.5 m higher above the
floor in vicinity of the Kaaba (Yousef, 1992).
Fig 1: location map of the study area.

The urban pattern of Makkah metropolitan city

is unique and controlled by the geomorphological
setting; the city has sprawled in radial direction on
the limited surface areas of the complex alluvial
channels separated by the steep mountain ranges. It is
estimated that approximately 9000 hectares of
alluvial channels and footslopes of the mountain were
converted into urban areas from 1978 to 2000 (Al-
Ghamdi and Al-Najjar, 2002). The development also
encroached the alluvial corridors, which dissect the
barrier mountain ranges, and thus the urban areas in
different subcatchments are locally connected.
Fig 2: A simplified geological map shows the main rock
Moreover, tunnels or rock-cut corridors are
units and the complex alluvial patterns, note the
commonly constructed in mountain ranges that extent of developed urban areas
obstacle the connectivity of neighborhood. The Holy
Mosque of Makkah (i.e. Al Al Masjed Al Haram) is As a result of the repeatability of flash floods
located within the lower reaches of Wadi Ibrahim several mitigation measures and different engineering
catchment; it embraces the Kaaba, and the sacred structures are constructed to collect and discharge the
well of Zamzam (located in a basement room 20 m runoff from the urban and mountaneous areas in
east of the Kaaba). The Zamzam well is about 30 m Wadi Ibrahim. An open channel with a maximum
deep and it taps the groundwater from the wadi cross-sectional area of 2 m2 constructed in the
alluviam (i.e. the upper half) and the underlying upstream zone of the catchment, in addition to

582
 Life Science Journal, 2012;9(1) http://www.lifesciencesite.com

several other interconnected lines of box culverts urban areas constructed in the pathways of these
(cross sections vary from 4.5 to 18 m2) constructed in thalweg drainage lines.
the middle and lower parts of the catchment. A new
culvert of 2.5 m depth and 3.5m width is being
constructed to convey the runoff water from the
urban areas within the upper sub-catchment
northward into neighboring Wadi Al-Ashr via an
alluvial channel, which breaches the barrier mountain
range (Saudi Geological Survey, 2011). However,
hazardous flash floods are developed and the most
recent event of 30th of December 2010 has produced
fatalities as large urban areas were flooded.
Therefore, it is necessary to assess the impact of
recent development in Wadi Ibrahim on the
hydrological balance and the adequacy of the
installed flash flood mitigation measures. Visual Fig 3: The total annual rainfall of Mecca area modified
interpretation of recent satellite images shows that after Yousef (1992).
most of the alluvial bed of Wadi Ibrahim has been
converted into urban areas, except few small and The DEM was also used to determine the
scattered parcels of bare alluvial soil in the supper spatially distributed time-area zones of the
sub-catchment. There is also a growing concern of catchment, which conforms the runoff hydrograph
the impact of these landcover changes on the when rainfall pattern and abstraction losses are
recharge and quality of the groundwater water within uniform across the catchment (Maidment, 1993). It is
Zamzam well surface catchment (Saudi Geological interesting to note that the application of pure flow
Survey, 2011). routing that neglects the volume of losses as the flow
is transmitted from one zone into another, was not
2. Data and methods: appropriate to the modeling of dryland hydrographs,
The morphometrical parameters of Wadi as it omits one dominant hydrological processes (i.e.
Ibrahim catchment was delineated from the available transmission loss into alluvium channels) (El
SRTM DEM following the ArcInfo multi-steps Bastawesy et al., 2009). Here in the alluvium
procedure of the D-8 method (Fig 3). The processing channels are no longer recognized and have been
of a DEM to produce hydrologically correct and converted into urban. Therefore the Maidment
connected drainage networks requires that sinks be technique of pure flow routing is acceptable; it will
first removed (Jenson and Dominique, 1988; Mark, be used to estimate the runoff hydrograph resulting
1984; Wise, 2000). The naturally occurring sinks are from the extreme rainfall events. Therefore, this
not common within the study area. Only few cells time–area diagram for the catchment can represent a
were missing elevation data (i.e. voids) resulting spatially distributed unit hydrograph without the need
from the generation technique of the SRTM DEM, for empirical functions for the time of concentration.
and thus they have been assigned elevation values The flow routing is simulated as a purely translation
from the neighbouring cells via the 'filling' step. A process that neglects abstractions due to storage or
given sink is filled to reach the nearest lowest loss over the flow pathway, and the runoff produced
elevation, the boundaries of the filled area may then from any time-area zone will reach outlet of the
be part of new sinks, which then need to be filled, catchment in a given time. Runoff will follow the
and this iteration process is repeated until all the pits same routes with the same velocity regardless of
are removed. Once a hydrologically connected DEM excess rainfall depth. The runoff velocities were
is produced by sink removal, flow direction is estimated using the Manning Equation (equation 1),
calculated for each cell in the DEM into the most which is widely been used in modeling the flows of
downslope of the neighbouring cells. The 'flow natural open-channels as well as storm-drainage from
direction' grid is then used to calculate a grid of 'flow urban areas (e.g. Ramier et al., 2011).
accumulation', and finally thresholding is applied to
extract up-slope contributing areas for all the V=(R^0.67*S^0.5)/n equation ( 1)
drainage channels in the DEM (Jenson and
Domingue, 1988). The resulting drainage networks Where V is the cross-sectional average velocity
were overlaid on the satellite images to check that (m s-1); n is the Manning coeffiecent of roughness; R
they correspond to visible wadis and to identify the is the hydraulic radius (m); S is the slope of the water
surface, which is assumed to be parallel to the slope

583
 Life Science Journal, 2012;9(1) http://www.lifesciencesite.com

of the channel bed. Municipal streets, whether changes of the landscape on the recharge to
planned or not, conveys large portion of runoff, and groundwater aquifer feeding the sacred well of
the flow- parameters of cross sectional areas (i.e. Zamzam. The analysis of recent satellite images and
width * height of the platform) and wetted perimeter field observations showed that urban development
were measured. The Manning's n was averaged and has sprawled on the limited alluvium soils and
set to 0.06 for hillslopes and 0.025 for urban areas; adjacent foot slopes of the mountains. Because the
these values are typical of reported values in estimated infiltration rates at the point-scale by the
literature (e.g. Mignot et al., 2006). Slopes were 'double ring ' are highly variable and associated with
estimated from the available DEM. Once these time- much uncertainly with up-scaling to the catchment
area zones was determined, the climatic data obtained level; the simulated hydrograph considered a
for Makkah area were then analyzed to estimate the uniformly distributed runoff coefficients estimated
designed storm that will be used in computing the from the opportunistic observations and
hydrograph for the catchment. A statistical measurements of the 30th of December 2010
probability technique was adopted; first the climatic hydroclimatical parameters. The discharge was
data was ranked (i.e. from highest to lowest), then the measured at a well-defined cross sectional area of the
return periods of the extreme storm was calculated trapezoidal concrete channel in the upper part of
using the following probability function (FAO, Wadi Ibrahim (Fig 5). The upstream flow
1991): contributing area at this channel cross section is
approximately 6.7 km2, the rainfall total depth was
P(%)=(M-0.375*100)/(N+0.25) equation ( 2) approximately 51 mm as measured at the nearest
meteorological station of The Costudian of the Two
Where; P = probability in (%) of the Holy Mosques institute for Hajj Research , and the
observation, M = the rank of observation, and N = the estimated runoff coefficient of this specific sub
total number of observation used. catchment is 60%. The catchment-simulated runoff
A satellite image of SPOT 5 acquired at 2010 hydrographs for the 30th of December flash flood
was also investigated to determine the landcover and event is shown in Fig (6). It is important to consider
hydrological parameters within the catchment (Fig 4). that the urban development on the bare-soil parcels in
the catchment will increase the runoff coefficient, as
the alluvial surfaces will be obscured and mostly
covered by urban features. Thus, the sensitivity of
runoff hydrographs simulated at different runoff
coefficients (i.e. for bare soil, hill slopes and urban
areas) was investigated , and also the scenario of
developing urban areas on these bare soil was tested
for the impact on hydrograph magnitudes. These
results show that the subtle changes in runoff
coefficient was of significant impact on the
magnitudes and attenuations of hydrographs (Fig 7).
However, the changes of runoff coefficient for bare
soil parcel were of the least impact on the developed
hydrographs; it may be because these alluvium soils
cover small surface areas of the whole catchment. In
contrary, in large-area catchments the hydrograph
(i.e. attenuation and duration) is mostly determined
by the runoff coefficient and transmission loss into
this alluvium (El Bastawesy et al., 2009). Frankly
Fig 4: The SRTM DEM for Wadi Ibrahim catchment, speaking, the flows developed within the upper
the location of The Holy Mosque is indicated by
black square.
reaches of large and undeveloped catchments may not
reach the outlet as they totally seep into the
3. Results: underlying alluvium while moving downstream along
Hydrology of the dry land catchment of Wadi these channels. In the catchment of Wadi Ibrahim
Ibrahim is modeled to assess the impact of the urban areas are developed on almost all of the
occasional flash floods hazard, and to address the alluvial channel beds and these changes of the land
negative impact of the adopted mitigation measures. cover has minimized the transmission loss. But on
It is so important to note, therefore the drastic the other hand, it has amplified the peaks of
hydrographs of the flash floods, and it thus increased

584
 Life Science Journal, 2012;9(1) http://www.lifesciencesite.com

the threat of flash floods to urban areas in the

downstream. Tens of vehicles were drifted during
the 30th of December 2010 flash flood event; and as a
result several people lost their lives. The torrential
flows were accompanied with loads of sediments
gullied from the remnant alluvial terraces hanging
upon some hill slopes as well as some rock boulders
of different sizes, which were dragged by flows from
the surrounding hill slopes and the artificially-cut
foot slopes areas. The sediment loads can reduce the
competent of the installed rainfall-sewage system,
which could be partially blocked by the trapped
sediments. However, the captured photos on 30th of
December 2010 clearly show that the capacity of the
culverts-discharge was exceeded and the flow depth
reached more than 50 cm on the main roads located
in the downstream of Wadi Ibrahim. Furthermore, the Fig 6: A field photo of the channeled flash flood of the
flood water has gushed into air from a manhole cover 30th of the 30th of December 2010 event.
of one of the sewage- culverts, which may have
contributed to increasing the surface runoff on
downstream urban areas (Fig 8). These undesired
outflows may have developed by the intense pressure
caused by excessive water than the designed-
discharge capacity of the culverts, or due to the
clogging of this culvert by washed debris at some
point downstream. Therefore, the likely development
of a flash flood from more intense rainstorm (i.e.
similar to those of 1969 and 1998), may be of serious
consequences, unless extra-mitigation measures are
considered. The current installed culverts and
rainfall-sewage system were not sufficient to control
Fig 7: The simulated hydrograph for the 30th of
the flows of flash floods during the latest event.
December flash flood event in Wadi Ibrahim.

Fig 8: The sensitivity of simulated hydrographs for

changes in the landuse and associate runoff
coefficients, note the increase of peak discharge
and the decrease of time to peak when runoff
coefficient is increasing.
Fig 5: The determined landcover units for Wadi
Ibrahim using a SPOT 5 satellite images of 2010.

585
 Life Science Journal, 2012;9(1) http://www.lifesciencesite.com

drift shrubs and wood lines) (e.g. Baker, 1987) is also

insufficient to represent the complex spatial and
temporal variability of flows and associated sediment
erosion pattern. Therefore, the developed
hydrographs of dryland catchments always contains
considerable uncertainty. Lots of research has
investigated the subtle sensitivity of the hydrological
model results to the minor changes in values of used
parameters, such as the empirical roughness
coefficient of Manning's widely used to estimate
channel-flow velocities (El Bastawesy et al., 2009).
Thus, the modeling of flash floods needs to be
further improved to minimize uncertainty in order to
better manage these torrential flows. Particularly for
urban areas and facilities that are being rapidly
developed within certain dryland catchments due to,
but not limited to the availability of local valuable
socioeconomic resources. In Wadi Ibrahim
catchment various mitigation measures are currently
adopted to control the impact of flash floods on the
Holy Mosque as well as other urban areas. The
efficiency of these installed measures was recently
tested as a severe flash flood has occurred on the 30th
of December 2010 following the precipitation of 51
mm on average. Unfortunately, the modeling
approach behind these implemented projects are not
available, and it is not known if these engineering
structures had considered the scenarios of landuse
and landcover change within the catchment for their
impact on altering the hydrograph as well as recharge
to the ground water aquifer of the area. It is clear that
the problems created by converting alluvial areas into
urban are complex, and the concept of 'complete'
flood control yet does not exist. In Wadi Ibrahim two
main methods are being adopted to control flash
floods, 1); Rainfall waters are collected through a
Fig 9: The outflows from sewers (top) and the runoff
hazards on municipal streets of Mecca during
sewage networks and confined subsurface culverts 2);
the 30th of December 2010 flash flood (bottom). Conveying flash floods from the upper part of Wadi
Ibrahim into Wadi Al Ashr via a subsurface culvert,
4. Discussion: which is being fed from a rainfall sewage system and
The hydrology of flash floods developed within an open-lined channel. It was noticed that a
the dryland catchment is usually not fully understood. considerable amount of flows during the latest event
The observations and in situ data measurements are were moving along the main municipal streets; thus
sparse; also the gauging of flash floods is very rare adequacy of the current rainfall-sewage system can
when compared with the seasonal or perennial flows be questioned. Obviously flash floods will continue
of more humid areas. As such the modeling of flash to be a threat for urban areas in Wadi Ibrahim unless
floods can be grossly inadequate, particularly when it a more competent management is addressed. Herein
is based on calculated empirical parameters or data it is suggested that several small-size dams have to be
obtained from very few rainfall and runoff gauging constructed at the bottom of the fingertip drainage to
stations. This data limitation may leads to erroneous retain flows and sediments. Consequently high-
conclusion, as the intricate meso-scale atmospheric velocity runoff will be reduced and values of peak
patterns and complex terrain processes that control discharge will be lessened. Moreover, these small
desert flash floods are neglected (Pilgrim et al., 1998; dams can act as a point-source recharge to the alluvial
Greenbaum et al., 1998). The alternative approach to aquifer, where most of the harvested water will seep
indirectly estimate flow discharge from the analyses into the underlying alluvium.
of paleostage indicator (i.e. slack water deposits and

586
 Life Science Journal, 2012;9(1) http://www.lifesciencesite.com

Of course further investigation is required for more severe rainfall events (e.g. The 1969 and 1998
this particular suggestion to assess the cost-benefit storms) are re-occurred. Indeed the flows of these
analysis, feasibility of construction and the severe storms will far exceed the maximum discharge
hydrological connectivity of the first-order drainage capacities of installed culverts, and will develop
segments to the main alluvial aquifer. In the mean considerable hazards. Furthermore, gullying of
time, the entire hydrological management of Wadi sediments from the abandoned fluvial terraces on
Ibrahim catchment needs a further assessment of the hillslopes can cause clogging problems to the
effluent pattern of rainfall-sewage downstream. rainfall-sewage and delimit their designed capacity. It
Indeed, the prevailing arid conditions and the severe is highly recommended to consider the construction
shortage of underground water resources in the area of several small dams at the fingertip channels, in
require a careful management of the fragile water order to retain considerable amount of water and
resources. The drainage pattern of Red Sea region sediments. These dams could be acting as a point-
(i.e. including Makkah area) is complex, structurally source recharge to the underlying alluvium aquifer, as
controlled and underwent different phases of the original replenishment of this aquifer from
morphotectonic evolution related to the Tertiary- channeled runoff was greatly reduced by urban
Quaternary motions in connection with Red Sea development.
Rifting ( Alwash and Zakir, 1992). As such the
alluvium thickness is highly variable along the Acknowledgment:
downstream profile of these wadis; several contact- This research has been supported by the Centre
springs used to discharge water along the transect of Research and Excellence in Hajj and Omrah
fault Plaines. The alluvial aquifers in Makkah area (HajjCore), Umm Al-Qura University, Makkah,
are mostly depleted by excessive pumping of the Saudi Arabia. Under project number, entitled (the
groundwater for local irrigation and potable water properties of hillslopes in Makkah and the holy
supplies (Alwash et al., 1986). The current outlets of places and their effects on transport and buildings by
the rainfall-sewage for Wadi Ibrahim as well as for using remote sensing and GIS techniques).
the neighboring urbanized catchments, which are
embracing the metropolitan city of Makkah, can be Corresponding author
extended to a specific alluvial channel. This Mohammed El bastawesy
proposed collection of the harvested runoff would Geography Department, Umm Al-Qura University,
maximize the benefit from severe rainfall storms for Makkah, Saudi Arabia
groundwater recharge. Ideally, the selected alluvial [email protected]
reach has to be underlain by ‘a trough’ that attains [email protected]
considerable thickness, and also it should be
hydrologically separate from contamination by other References:
waste dump-sites and sanitary sewage outlets. Abou El-Magd, I. H., Hermas, E. A., Bastawesy, M.
2010. Integration of hydrological models and GIS
Conclusion: for the assessment and prediction of natural hazards
The problem of flash floods in Wadi Ibrahim in Abu Dabbab area, Red Sea region, Egypt. The
catchment is prominent, and it has been further Egyptian Journal of Remote Sensing and Space
complicated by the sprawl of urban development on Sciences, 13, 81 - 88.
the alluvial channels. Consequently, the significant Al-Ghamdi, S. A., and Al-Najjar, Y. M., 2002.
reduction of transmission losses into the underlying Analysis of urban growth and its patterns using
alluvium has increased the runoff coefficient and remotely sensed data: The case study of the Holy
discharge from this catchment. The adopted City of Makkah Al-Mukkramah (1987-2000).
management strategies for the flash flood included Journal of King Abdul-Aziz University, Special
the construction of a rainfall-sewage system in issue of the 20th Anniversary of King Fahd
addition to conveyance culverts in the upper part of Coronation, pp231-273.
the catchment to transfer the flows into Wadi Al Ashr Almazroui, M., 2011. Calibration of TRMM rainfall
(i.e. a hydrologically separate drainage basin). The climatology over Saudi Arabia during 1998–2009.
latest flash flood event of the 30th of December 2010 Atmospheric Research, 99, 400-414.
clearly showed the insufficiency of the current Alwash, M.A. Zaidi, S.M.S. and Terhalle, U., 1986.
mitigation measures to deal with flash floods; the Description of arid geomorphic features using
main municipal streets were flooded and fatalities landsat-TM data and ground truth information
occurred as several vehicles were drifted. It is evident (Wadi Fatima, Kingdom of Saudi Arabia). Catena,
that the urban areas of Wadi Ibrahim are prone to 13, 277-293.
more serious threat of flash floods, particularly if

587
 Life Science Journal, 2012;9(1) http://www.lifesciencesite.com

Alwash, M.A., and Zakir, F.A.R., 1992. Tectonic Hossain, F., Anagnostou, E. N., and Dinku, T.,
analysis of the Jeddah Taif area on the basis of 2011. Sensitivity Analyses of Satellite Rainfall
LANDSAT satellite data. Journal of African Earth Retrieval and Sampling Error on Flood Prediction
Sciences (and The Middle East), 15,293-301. Uncertainty. IEEE Transactions On Geoscience and
Baker, V. R. 1987. Paleoflood hydrology and Remote Sensing, 42, 130-139.
extraordinary flood events. Journal of Hydrology, Huffman, G.J., Adler, R. F., Bolvin, D. T., Gu, G.,
96, 79-99. Nelkin, E. J., Bowman, K. P., Hong, Y., Stocker, E.
Band, L. E. 1986. Topographic partition of watershed F., Wolff, D. B., 2007. The TRMM Multi-satellite
with digital elevation models. Water Resources Precipitation Analysis: Quasi-Global, Multi-Year,
Research, 22, 15-24. Combined-Sensor Precipitation Estimates at Fine
Borga, M., Gaume, E., Creutin, J. D., and Marchi, L., Scale. Journal of Hydrometeorology, 8, 38-55.
2008. Surveying flash floods: gauging the ungauged Jenson, S. K., Domingue, J. O., 1988. Extracting
extremes. Hydrological processes, 22, 3883-3885. topographic structure from digital elevation data for
Chin, A., Gregory, K.J., 2001. Urbanization and geographical information system analysis.
adjustment of ephemeral stream channels. Annals Photogrammetric Engineering and Remote Sensing,
of the Association of American Geographers 91, 54, 1593–1600.
595–608. Leveson, D., 1980. Geology and the urban
Chorowicz, J., Ichoku, C., Riazanoff, S., Kim, Y. and environment. Oxford University Press, New York,
Cervelle, B. 1992. A combined algorithm for 386pp.
automated drainage network extraction. Water Maidment, D. R., 1993. Developing a spatially
Resources Research, 28, 1293-1032. distributed unit hydrograph by using GIS. In:
Cooke, R., Brunsden, D., Doornkamp, J., Jones, D., Proceeding of the Vienna Conference on
1982. Urban Geomorphology in Drylands. Oxford Application of geographic Information Syaten in
University Press, Oxford. Hydrology and Water Resources, IAHS, 211, 181-
El bastawesy, M., 2007. Influence of DEM source 192.
and resolution on the hydrographical simulation of Mark, D M., 1984. Automatic detection of drainage
Wadi Keed catchment, Sinia, Egypt. Egyptian networks from digital elevation models.
Journal of Remote Sensing and Space Sciences, 10, Cartographica, 21, 168-178.
53-62. Mignot, E., Paquier, A., Haider, S., 2006. Modeling
El Bastawesy, M., White, K., and Nasr, A., 2009. floods in a dense urban area using 2D shallow water
Integration of remote sensing and GIS for equations. Journal of hydrology,327, 186-199.
modelling flash floods in Wadi Hudain catchment, Milewski, A., Sultan, M., Yan, E., Becker, R.,
Egypt. Hydrological Processes, 23, 1359-1368. Abdeldayem, A., Soliman, S., Abdel Gelil, K.,
El Bastawesy, M., Faid, A., and El Gammal, E., 2009. A remote sensing solution for estimating
2010. The Quaternary development of tributary runoff and recharge in arid environments, Journal
channels to the Nile River at Kom Ombo area, of Hydrology, 373, 1-14.
Eastern Desert of Egypt, and their implication for Milzow, C., Kgotlhang, L., Kinzelbach, W., Meier, P.
groundwater resources. Hydrological processes, 24, and Bauer-Gottwein, P. 2008. The role of remote
1856-1865. sensing in hydrological modelling of the Okavango
FAO. 1991. Water Harvesting. A manual for the Delta, Botswana. Journal of Environmental
design and construction of water harvesting Management, 90, 2252-2260.
schemes for plant production, Rome, Italy. Pilgrim, D. H., Chapman, T. G., and Doran, D. G.
Foody, G.M., Ghoneim, E M., and Arnell, N.W., 1988. Problems of rainfall-runoff modelling in arid
2004. Predicting locations sensitive to flash and semiarid regions. Hydrological Science Journal,
flooding in an arid environment. Journal of 33, 379-400.
Hydrology, 292, 48–58. Ramier, D., E. Berthier, E., and Andrieu, H., 2011.
Gheith, H., Sultan, M., 2002. Construction of a The hydrological behaviour of urban streets: long-
hydrologic model for estimating Wadi runoff and term observations and modelling of runoff losses
groundwater recharge in the Eastern Desert, Egypt. and rainfall–runoff transformation. Hydrological
Journal of Hydrology 263, 36–55. Processes, 25, 2161-2178.
Greenbaum, N., Margalit, A., Schick, A. P., Sharon, Rendell, H., Alexander, D., 1979. Note on some
D., Baker, V.R., 1998. A high magnitude storm spatial and temporal variations in ephemeral
and flood in a hyperarid catchment, Nahal Zin, channel form. Geological Society of America
Negev Desert, Israel. Hydrological Processes, 12, Bulletin 90, 761–772.
1-23.

588
 Life Science Journal, 2012;9(1) http://www.lifesciencesite.com

Saudi Geological Survey, 2011. Wadi Ibrahim Tooth, S., 2000. Process, form, and change in dryland
Environmental Management System (WIEMS), rivers: a review of recent research. Earth-Science
Progress report 6, pp 39. Reviews 51, 67–107.
Schultz, G. A., 1987. Parameter determination and Yousef, B. E. M. A., 1992. The climate of Makkah
input estimation in rainfall-runoff modeling based area. Journal of Umm Al-Qura Social Sciences, 15,
on remote sensing techniques. Water for the Future: 1-94.
Hydrology in Perspective (Proceedings of the Rome Walling, D.E., Gregory, K.J., 1970. The
Symposium, April 1987). IAHS Publication number measurement of the effects of building construction
164. on drainage basin dynamics. Journal ofHydrology
Scipal, K., Scheffler, C. and Wagner, W., 2005. Soil 11, 129–144.
moisture-runoff relation at the catchment scale as Wise, S. M. 2000., Assessing the quality for
observed with coarse resolution microwave remote hydrological applications of digital elevation
sensing. Hydrology and Earth System Sciences, 9, models derived from contours. Hydrological
173-183. Processes, 14, 1909-1929.
Shang, J., Wilson, J. P., 2009. Watershed Wolock, D. M. and Price, C. V. 1994. Effects of
urbanization and changing flood behavior across digital elevation map scale and data resolution on a
the Los Angeles metropolitan region. Natural topographically based watershed model. Water
Hazards 48, 41–57. Resources Research, 30, 3041-3052 .
Smith, D. P., Bedient, P. B., 1981. Preliminary model Zhang, W. and Montgomery, D. R. 1994. Digital
of an urban floodplain under changing landuse. elevation model grid size, landscape representation,
Journal of Hydrology, 51, 179-185. and hydrological simulations. Water Resources
Suriya, S., and Mudgal, B. V., 2011. Impact of Research, 30, 1019-1028.
urbanization on flooding: The Thirusoolam sub
watershed – A case study. Journal of Hydrology ( in
press).

2/8/2012

589