Geocarto International

ISSN: 1010-6049 (Print) 1752-0762 (Online) Journal homepage: http://www.tandfonline.com/loi/tgei20

Rainfall-runoff Modeling and Water Balance

Analysis for Al-Hindiyah Barrage, Iraq Using
Remote Sensing and GIS

Hussein Sabah Jaber, Shattri Mansor, Biswajeet Pradhan & Noordin Ahmad

To cite this article: Hussein Sabah Jaber, Shattri Mansor, Biswajeet Pradhan & Noordin Ahmad
(2016): Rainfall-runoff Modeling and Water Balance Analysis for Al-Hindiyah Barrage, Iraq Using
Remote Sensing and GIS, Geocarto International, DOI: 10.1080/10106049.2016.1213889

To link to this article: http://dx.doi.org/10.1080/10106049.2016.1213889

Accepted author version posted online: 14

Jul 2016.
Published online: 14 Jul 2016.

Submit your article to this journal

Article views: 4

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=tgei20

Download by: [University of Exeter] Date: 19 July 2016, At: 02:22

 Publisher: Taylor & Francis

Journal: Geocarto International

DOI: http://dx.doi.org/10.1080/10106049.2016.1213889

Rainfall-runoff Modeling and Water Balance Analysis for Al-Hindiyah

Barrage, Iraq Using Remote Sensing and GIS
Downloaded by [University of Exeter] at 02:22 19 July 2016

Hussein Sabah Jaber*1,2, Shattri Mansor1, Biswajeet Pradhan1 and Noordin Ahmad1

1
Department of Civil Engineering, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor, Malaysia
2
Department of Surveying Engineering, University Baghdad, Aljadriyah, Baghdad, Iraq

Corresponding Author: [email protected], Phone NO. 0060182383781

 Rainfall-runoff Modeling and Water Balance Analysis for Al-Hindiyah
Barrage, Iraq Using Remote Sensing and GIS
Downloaded by [University of Exeter] at 02:22 19 July 2016

Abstract

Investigation of rainfall-runoff modeling is an important subject to develop any available

means to water supply, which maintains human life such as runoff harvesting method.
This study aims to analyze and understand the rainfall-runoff relationship in a part of
Babil city, Iraq. Curve Number (CN) which is a function of land use, soil texture, soil
moisture and land slope is used in this study. Remote sensing and GIS are used to
analyze the data and to produce the runoff depth map for the study area. Then the runoff
depth is used with rainfall to investigate the relationship between them using linear
correlation. This study showed a strong linear relationship between rainfall and runoff
(R2 =0.992). It indicates that in the absence of rainfall data, runoff data can be used to
estimate rainfall amount. Also, the study revealed through water balance analysis that the
average monthly change in storage.

Keywords: runoff-rainfall modeling, curve number, water balance, potential runoff coefficient

1. Introduction

Runoff is the flowing of precipitated water in the catchment area through a channel after
satisfying all surface and sub-surface losses (Schultz, 1994). Accurate estimation of runoff from
 rainfall is important for land and water resource management. Arid and semi-arid regions defined
as areas where water is at its most scarce. The hydrological regime in these areas is extreme and
highly variable, and they face great pressure to deliver and manage freshwater resources.
Rainfall-runoff relationships are very important for the catchment management and the
sustainable development of the water resources .Rainfall is one of the essential hydro-climatic
forcings in the modeling of basin systems (Masoud, 2014; Vaze et al., 2010). The rainfall–runoff
relationship is strongly dependent on soil, vegetation, and topographic characteristics of the
catchment. The water cycle and water resources have assumed key worldwide significance.
Particularly in semi-arid climatic environments such as Al-Babil city, where the hydrological
processes are widely variable both in time and in space (Isaev and Mikhailova, 2009). Therefore,
Downloaded by [University of Exeter] at 02:22 19 July 2016

the quantification of renewable water resources in countries such as Iraq is of paramount

importance for hydrological and environmental planning (O'Loughlin et al., 1996; Taguas et al.,
2015).

Geographic Information System (GIS), which designed to restore, manipulate, retrieve and
display spatial and non-spatial data, is an important tool in analysis of parameters such as land
use, land cover, soils, topographical and hydrological conditions (Cheng et al., 2006). In
addition, GIS has become a critical tool in hydrological modeling because of its capacity to
handle a large amount of spatial and attribute data .Some of its features, such as map overlay and
analysis, help for deriving and aggregating hydrologic parameters from different sources such as
soil, land cover, and rainfall data (Baig et al., 2008; Mahmoud, 2014). Remote sensing along
with GIS application help to collect, analyze and interpret the multidisciplinary data rapidly on
large scale and is very helpful for watershed planning. Estimation of runoff potential from
ungauged watersheds using conventional methods requires much time and efforts. Conventional
methods of runoff measurements are not easy for inaccessible terrain and not economical for a
large number of small watersheds (Deshmukh et al. (2013). Remote sensing and GIS can
augment the conventional method largely in rainfall-runoff modelling (Cheng et al., 2006; Liu,
2004).

There are several methods available for estimation of runoff. Among them, the Natural
Resources Conservation Services Curve Number (NRCS-CN) is most widely used method
because of its simplicity, flexibility and versatility for small areas (Deshmukh et al., 2013).
 Deshmukh (Deshmukh et al., 2013) demonstrated an integrated remote sensing and GIS based
methodology for estimation of runoff in Burigana watershed. In addition, (Mahmoud, 2014) used
the same method to determine the potential runoff coefficient, runoff depth, and runoff modeling
for Egypt using GIS, available data and remote sensing. Furthermore, the results of a study by
(Melesse & Shih, 2002) showed that Landsat images were helpful in identifying the runoff
response of watersheds regardless of their somewhat varied spatial resolution. On the other hand,
a grid or cell based process oriented distributed rainfall–runoff model capable of handling the
catchment heterogeneity in terms of distributed information on landuse, slope, soil and rainfall is
developed and applied to isolated storm events in several catchments (Blanco et al., 2013; Jain et
al., 2004). Moreover, (Gad, 2013) developed an automated rainfall-runoff model for engineering
Downloaded by [University of Exeter] at 02:22 19 July 2016

applications in semi-arid regions. In the recent past, the use of Artificial Intelligences (AI)
techniques, e.g., Artificial Neural Networks (ANNs), Adaptive Neuro-Fuzzy Inference System
(ANFIS) and Genetic Programming (GP) in water resources engineering has become viable. So
far, numerous works have been reported in the literature regarding the application of ANNs in
modeling rainfall-runoff and other hydrologic factors (Kisi et al., 2013).

Thus, this study explores the relationships between the runoff and rainfall in the Al-Hindiyah
barrage, Iraq. Further, it analyzes the water balance for the catchment to evaluate the amount of
yearly runoff in the catchment. The information provided by water balance analyzes is important
to understand the characteristics of the study catchment. In addition, the water balance model
provides information about the amount of water is available for irrigation and other purposes.

1.1 Data Source

Various data sources are used in this study to produce final runoff depth map for the study area.
Some of these data are downloaded free of charge from internet and others are provided by local
organizations and institutions in Iraq. Table 1 describes the list of data used in this study with the
source of each dataset.

1.2 Study Area

The study area is located in Babil province in Iraq. It extends between longitudes (44° 00' to 44°
40' E) and latitudes (32° 30' to 33°00' N) as shown in Figure 1. The area bounded amounts to
(1723 km2). The land is flat topography with average elevation of 38.00 m above mean sea level.
 Also, it consists of four types of land covers: water bodies, bare soil, urban area and agriculture
land. These data were gathered from Ministry of Agriculture and Irrigation, Iraq.

In general, the topography of the study area forms of two figures namely, the shoulders of natural
rivers formed by the sedimentation processes during floods that were exposed to the province.
Babil province is one of the most suitable surface section of the agricultural terms due to the
quality of the soil and lowers the water table. The soil of the study area consists of soft granules
and mudslides in the mostly poor permeability and drainage (Al-Ansari et al., 2015).

Cropping patterns indicate the kind and sequence of crops grown over a period of time on a
given area of land affected by significant factors, such as climate, topography, and soil. Actually,
Downloaded by [University of Exeter] at 02:22 19 July 2016

the cropping pattern policy in the study area is not clear because of weakness of planning tool
and poor consideration for available water resource. However, the climate plays the main
important factor to determine the quality and intensity of natural vegetation.

The area of study is characterized in three cropping seasons, winter, summer, and perennial. The
main winter crops are wheat, barley, and winter vegetable. Maize and summer vegetables are the
dominant summer crops, while the orchards, palms and alfalfa are the dominated perennial crops.
In addition, there are two main types of soil in this study area: loam and sand. Table 2 shows
general soil data, such as soil moisture, infiltration rate and soil moisture depletion, these data
have been gathered from (Allen et. al., 1998) and (Ministry of Water Resources, Iraq, 2014).

Temperatures rates over the study area generally characterized by a low annual rates in the
winter to hit in the month of January to (10.5 C°) then begin to rise gradually to reach the highest
in July and August to (34.7 C°) for both months and then resume the decline trend in autumn and
winter. Rainfall is scarce in the central zone of Iraq. It irregular and unpredictable. The
maximum amount of total rainfall takes place in January which it is only 21.50 mm. whereas the
minimum amount of total rainfall takes place in July which it is closed to zero mm. Annual
rainfall ranges from 140 mm to 180 mm (Al-Ansari et al., 2015) .
 2. Methodology

Both remote sensing and GIS techniques are used to process the raw data to investigate the
rainfall- runoff modeling for a part of Al-Babil province in Iraq. Remote sensing is used to
produce the land cover map for the study area using image classification technique for Landsat
OLI data, where OLI is abbreviated of words (Operational Land Imager). OLI band data can
convert digital number to TOA planetary reflectance, a sensor of Landsat receiving 16 bit where
an especial equation can be used to convert digital number (DN) values to TOA reflectance for
Operational Land Imager OLI data (Landsat 8, 2014).

GIS is used for producing slope map by applying slope algorithm on DEM data, where DEM is
Downloaded by [University of Exeter] at 02:22 19 July 2016

abbreviated for words (Digital Elevation Model) and it is a vital tool for initial upstream
exploration operations. It is used in many fields such as extracting terrain parameters for
geomorphology and modeling water flow for hydrology (Janssen, 2000). GIS is also used to
produce final soil map for the study area by overlapping the soil layers and clipping for the study
area extent. Raster calculator is then used in GIS to apply the equations of potential runoff depth
that depending on Curve Number (CN) method and these equations were explained in section
(2.3.1) (Masoud, 2014) and (USAD, 2004) to produce the final map of runoff depth for the study
area. The overall research activities are described in the Figure 2.

2.1 Preparation of soil texture in GIS

The soil texture of the study area was downloaded from the USGS website and then refined
using the Iraqi earth exploratory map. Generally, the study area has seven types of the soil
textures; river levee, basin depression, gypsiferous gravel, poorly drained periodically flooded,
active dune and river basin, but the most of the study area (middle and north parts) has two main
soil classes: loam and sand as shown in the Figure 3. The legend of the original soil map of the
World (FAO, 1974) includes an estimated 4930 different map units, which involve soil units or
associations of soil units. When a map unit is not homogeneous, it is composed of a dominant
soil and component soils. The latter are: associated soils, covering at least 20% of the area; and
inclusions, important soils which cover less than 20% of the area.
 The soil map of the study area was classified into two main classes: sandy and loamy soil. The
loamy soil which has a moderate infiltration rate when is thoroughly wetted was classified as
mainly or moderately deep infiltration, moderately to-well drained soils with moderately fine to
moderately coarse textures. Sandy soil has the highest infiltration rate. These findings will cause
variations in estimation runoff values because the soil texture is varied.

2.2 Derivation of land use/land cover classification map

Land cover map of the study area is produced using Landsat OLI data downloaded from USGS
website on 21-March-2014. First, the data was processed in ENVI (Environmental Visualize
Downloaded by [University of Exeter] at 02:22 19 July 2016

Image) software to rectify the image to the proper coordinate system as same as with other layers
where twenty five sample points (Ground Control Points) were collected in a clearly delineable
locations on the images. These points (GCPs) were then used with the nearest neighbor
resampling method to rectify the images (Jensen, 2005).Then the process of radiometric and
atmospheric corrections was applied to produce top of atmosphere (TOA) image. In terms of
radiometric correction, information about the gain and offset found in the metadata was used to
convert the image digital numbers into radiance values.

Unfortunately, noise (error) can enter the data collection system at several points. For example,
radiometric error in remotely sensed data may be introduced by the sensor system itself when the
individual detectors do not function properly or are improperly calibrated. Second, the
intervening atmosphere between the terrain of interest and the remote sensing system can
contribute so much noise that the energy recorded by the sensor does not resemble that which
was reflected or emitted by the terrain (Jensen, 1996).

On the other hand, it is usually necessary to atmospherically correct the image if biophysical
parameters are going to be extracted from water bodies (e.g., chlorophyll, sediment, temperature)
or vegetation (e.g., biomass, leaf area index). The images were atmospherically corrected using
Dark Object Subtraction model. Dark object subtraction searches each band for the darkest pixel
value. Assuming that dark objects reflect no light, any value greater than zero must result from
atmospheric scattering. The scattering is removed by subtracting this value from every pixel in
 the band. This simple technique is effective for haze correction in multispectral data, but it
should not be used for hyper spectral data. (Jensen, 2005).

This data then was used with the field training data to produce land cover map using maximum
likelihood classification technique. The Kappa accuracy obtained from this processing is (83.2).
The final data product (see Figure 4) was exported to ArcGIS software to generate curve number
map for the study area. The study area covers by four main types of land; agriculture, bare soil,
urban and water bodies. The maturity of the land cover in the study area is used for agriculture
with around (80%).

2.3 Derivation of land slope

Downloaded by [University of Exeter] at 02:22 19 July 2016

SRTM data are used to produce slope map for the study area, which is required to correct the
curve number map to reduce the effect of land slope. 3D analysis of ArcGIS was used for
derivation of land slope by applying slope algorithm on DEM data. The DEM analyzed to remove
sinks and flat areas to maintain continuity of flow to the catchment outlets. The GIS used for
DEM preparation by filling the sink areas so the DEM is ready for the subsequent steps. The
output of the processing resulted in slope layer with percent unit. In this study, slope layer was
generated with percent unit to be used directly for correcting the curve number layer. Figure 5
shows the DEM and derived slope layer for the study area.

2.3.1 Formulation of runoff modeling equation

The mathematical form (equation 1) of the NRSC runoff equation describes the ratio of
amount of actual retention to watershed storage is equal to the ratio of actual direct runoff
to the effective rainfall as mentioned by (Masoud, 2014; Mockus, 2010).

= 1
−

where F= Actual Retention (mm), S = Retention (mm), Q = Runoff Depth (mm), P = Total
Rainfall (mm) and I = Initial Abstraction (mm).

The conservation of mass equation can be expressed in the form of water balance equation as
below in (equation 2).
 = − − 2

Substituting equation 2 in equation 1, results (equation 3):

− − = 3
−

Solving for runoff depth (Q) results in the runoff equation (equation 4)

−
= + 4
−

A relationship between S and I was developed by analyzing rainfall-runoff data for many small
Downloaded by [University of Exeter] at 02:22 19 July 2016

watersheds. This relationship had been made to eliminate the necessity of estimating both
parameters I and S. Equation (5) expresses this relationship.

= 0.2 5

Initial abstraction factor: The initial abstraction factor I is empirically derived from the
maximum soil water retention (Sn), which is related to the soil drainage characteristics (e.g., CN
values). I accounts normally for losses due to evaporation, plant uptake, and water retained in
surface depression during the rainfall event. Sn accounts for the total amount of water retained in
the drainage basin during the rainfall event, essentially I + Infiltration (Masoud, 2014; Mockus,
2010).

The final runoff depth equation is formed by substituting equation (5) in equation (4).

− 0.2
= ≥ 0.2 6
+ 0.8

Where S is retention parameter which varies with changes in land uses, soil texture and land
slope.it can be calculated from the equation (7) which is a function of CN.

1000
= − 10 7
 CN is the potential runoff coefficient or curve number. It can be estimated based on hydrologic
soil group (HSG) and land use. CN values range between 1 and 100. Higher the value of CN
indicates higher the surface runoff.

In order to produce potential runoff coefficient and runoff depth maps in GIS, the fundamental
equations described above were used. First, digital elevation model is used to generate slope map
for the study area. Then, soil map, land cover land use, and average precipitation are papered for
next step. The next step is to use soil map to estimate the hydrological group for soil types. After
that, LULC, soil, and precipitation layers are used to produce curve number map based on
estimated soil groups and LULC types. Then, the CN layer is corrected to the effects of terrain
slopes using LULC layer. Finally, equation (6) above is applied using Raster Calculator to
Downloaded by [University of Exeter] at 02:22 19 July 2016

produce the runoff depth map for the study area.

2.4 Average annual rainfall

The average precipitation data was collected from the Al-Hila metrological station located in the
study area as excel file. Based on the data that have collected, point shapefile is created in
ArcGIS with the attributes of rainfall data. Spline interpolation then used to create a rainfall
raster for the study area. Finally, the interpolated raster and other important layers are combined
to produce the final precipitation map as shown beside in Figure 6.

3. Results and discussion

Runoff curve number has been estimated based on the area's hydrologic soil group, land use, and
slope, in this study. First, the soil map was extracted from the global soil map of USGS and
further refined using the Iraqi expletory map in GIS environment. Then, the Landsat OLI data
was used for land use/land cover mapping. Support vector machine was used to classify the
Landsat image using training data randomly selected based on Google map. After that, the land
slope was derived from the SRTM derived digital elevation model (30m spatial resolution).
Next, the land use/land cover and soil texture information was used to derive the curve number
of the study area. The curve number then was adjusted using the land slope for accurate rainfall-
runoff modeling. Finally, the relationship of rainfall and runoff was described as a linear
relationship. The next subsequent sections describe the findings of this study.
 3.1 Potential runoff coefficient

Utilization of GIS and RS was conducted by Shereif et al. (Mahmoud et al., 2013) to determine
the potential runoff coefficient in Al-Baha region, Saudi Arabia. They reported that both of soil
map and land cover/use map were added by using the satellite imagery of the Al-Baha region. In
present study, GIS was used to combine the land use, slope, and soil maps into one map, and
then, a new field was added to the attribute table of the CN values. Figure 7a shows the runoff
coefficient derived in GIS using the mentioned maps and shows the values of CN for the study
area were varied from (64 to 100) where CN is the potential runoff coefficient or curve number.
It can be estimated based on hydrologic soil group, slope and land use. CN values range between
1 and 100 and higher the value of CN indicates higher the surface runoff in the study area.
Downloaded by [University of Exeter] at 02:22 19 July 2016

In addition, the Figure 7b shows the adjusted runoff coefficient, which was generated in GIS to
reduce the effect of land, slope on the estimating curve number from remote sensing data. The
minimum curve number of (64 and 69) was observed for the normal and adjusted curve number
respectively. This adjusting has been done by using land use layer and the CN value was
increased after adjusting process to enhance calculations and get better results where it ranged
from( 69 to 100 ). This variation in runoff curve number values because the soil texture is varied.
From this result it can be showed that loam soil is more suitable for runoff harvesting because of
its lowest infiltration potential. The runoff coefficient derived in this step was then used for
runoff –rainfall modeling.

3.2 Spatial distribution of runoff

This section presents the results of the rainfall-runoff modeling over the study area. Monthly
runoff depth is derived for this study as opposed to event runoff coefficient for monthly rainfall
since establishing the runoff amount could be available for agricultural production. This method
takes into account rainfall events that do not contribute significantly to any runoff and it used for
different purposes. Criteria that used by Ramakrishnan et al. (Ramakrishnan et al., 2009) was
runoff potential to select useful sites for different RWH/recharging structures in the Kali
watershed, Dahod district of Gujarat of India by GIS and remote sensing techniques. Monthly
runoff depth is derived using the monthly rainfall surplus and runoff coefficient per pixel using
 the raster calculator tool in ArcGIS, which gives an indication of the rainfall percentage that, is
transformed to runoff. The results of spatial distributions of modeled monthly runoff depth in
mm are shown in Figure 8. A variation from as low as 12.3 mm into a maximum of 13.7 mm was
observed due to divergence in topography and climate in the study area, where the largest rainfall
was observed in the southern part, while the lowest runoff was observed in the north of the study
area. From this method, total catchment runoff yearly was predicted, where value of maximum
catchment runoff for the studied area was (13.7064*12= 164.476) mm. These information
provided to know how much water is available in the catchment for the irrigation and other
purposes (domestic, industrial, etc.). Also, it can use for water balance analysis later. This
method was useful for Iraq because of there is a limited studied were done up to date in this area.
Downloaded by [University of Exeter] at 02:22 19 July 2016

3.3 Rainfall-Runoff relationship

Plotting catchment runoff coefficients provides a quick visual check of the robustness of runoff
estimates and can assist identify outliers. The relationship of runoff-rainfall shown in Figure 9
gives an indication of how the satellite image derived rainfall amounts fit to the gauge measured
rainfall. An indication that the absence of reliable ground measurements of rainfall product can
satisfactorily be applied to estimate the spatial rainfall distribution based on the value of (R2=
0.992) as obtained by linear modeling from Equation (8). Despite the high correlations, one of
the key problems may be the accuracy of the measured rainfall and its reliability (Senay &
Verdin, 2004).

= 0.9897 × 8

Goodness of fit: = 0.992. This means that as rainfall amount increase the actual retention
(rainfall depth-runoff depth) approximates a constant value, the maximum potential retention
(Mockus, 2010; Ponce & Hawkins, 1996).

3.4 Water balance analysis

To evaluate the water resource allocation for each planting season, the water balance equation
was used to find the changes in storage (ΔS) based on the curve number determined previously
(section 3.2). The monthly and annual water balances were calculated using the equation 9 as
following (Kato et al., 2013):
 = + − − 9

Where ΔS is the change in the amount of water stored in the surface layer (root zone), which
includes percolation and exploitation of groundwater (mm month-1), P is the precipitation, (mm
month-1), Ir is the irrigation intake (mm month-1), Q is the discharge to the river, (mm month-1),
and ET is the evapotranspiration (mm month-1).

Table 3 shows the average monthly change in storage and related parameters for the studied
catchment area. ΔS for the total study area on October 2014 was observed the lowest to be (-
218.14 mm) and on July 2014, the number was (22716.41 mm) as its highest value. The negative
values of ΔS indicated the water shortage(exploitation of groundwater )and the positive values of
Downloaded by [University of Exeter] at 02:22 19 July 2016

ΔS means that water was available for irrigation(percolation of groundwater).This emphasis that
there was a lot of amount of ground water in the studied area that it can used for irrigation
source.

4. Conclusion

It was established successfully in this study that remote sensing and GIS can provide the
appropriate platform for convergent of large volume of multi-disciplinary data. Besides, Remote
sensing data help a lot in SCS-CN model as it serve as input for determination of drainage
pattern, delineation of catchment, land use/land cover, etc., which when integrated with GIS can
be used in water resource management effectively. In this study, data from remote sensing and
GIS spatial analysis tools were used to establish a linear relationship between rainfall and runoff
in Al-Hindiyah barrage, Iraq. In addition, the study further explored the monthly distribution of
water balance.

The results of the study show that the runoff values in AL-Hindiyah barrage can be studied for
reliable accuracy along with the spatial variation of soil type and land-use type. The accuracy of
the linear relationship between rainfall and runoff of (R2=0.992) was achieved. This study
provides possibility to simulate the runoff either at a daily, monthly, seasonal, or annual scale,
for each rainfall or each storm event. This work can be helpful for estimating runoff at places
where observed runoff records are not available.
 In general, the runoff-rainfall modeling carried out in this study revealed the possibility of
predicting runoff flow in the study area using the field measured rainfall. On the other hand, the
study showed that the average monthly change in storage and related parameters for the studied
catchment area. ΔS for the total study area on October 2014 was observed to be (-218.14 mm)
and on July 2014, the number was (22716.41 mm). The negative values indicate the water
shortage and the positive values means that water was available for irrigation.

Acknowledgements

The authors like to thank all the staffs worked for producing this paper and enhanced the final
visualization of the maps. They also want to thank the ministry of higher education of Iraq for
Downloaded by [University of Exeter] at 02:22 19 July 2016

providing scholarship for the author to study his PhD at Universiti Putra Malaysia.

References
Al-Ansari, N., Ali, A. A., & Knutsson, S. (2015). Iraq water resources planning: Perspectives
and prognoses. Proceedings of the ICCCE, , 2097-2108.

Baig, M. A. A., Mallikarjuna, P., & Krishna Reddy, T. (2008). RAINFALL—RUNOFF

MODELLING—A CASE STUDY. ISH Journal of Hydraulic Engineering, 14(2), 18-35.

Blanco, C., Santos, S., Quintas, M., Vinagre, M., & Mesquita, A. (2013). Contribution to
hydrological modelling of small amazonian catchments: Application of rainfall–runoff
models to simulate flow duration curves. Hydrological Sciences Journal, 58(7), 1423-1433.

Cheng, Q., Ko, C., Yuan, Y., Ge, Y., & Zhang, S. (2006). GIS modeling for predicting river
runoff volume in ungauged drainages in the greater toronto area, canada. Computers &
Geosciences, 32(8), 1108-1119.

Deshmukh, D. S., Chaube, U. C., Hailu, A. E., Gudeta, D. A., & Kassa, M. T. (2013). Estimation
and comparision of curve numbers based on dynamic land use land cover change, observed
rainfall-runoff data and land slope. Journal of Hydrology, 492, 89-101.
 Gad, M. A. (2013). A useful automated rainfall-runoff model for engineering applications in
semi-arid regions. Computers & Geosciences, 52, 443-452.

Jain, M. K., Kothyari, U. C., & Raju, K. G. R. (2004). A GIS based distributed rainfall–runoff
model. Journal of Hydrology, 299(1), 107-135.

Kato, T., Kuroda, H., & Yoshida, K. (2013). GIS analysis of conjunctive water resource use in
nganjuk district, east java, indonesia. Paddy and Water Environment, 11(1-4), 193-205.

Kisi, O., Shiri, J., & Tombul, M. (2013). Modeling rainfall-runoff process using soft computing
techniques. Computers & Geosciences, 51, 108-117.
Downloaded by [University of Exeter] at 02:22 19 July 2016

Liu, Y. (2004). Development and application of a GIS-based hydrological model for flood
prediction and watershed management. Vrije Universiteit Brussel Press, Brussel, 315

Mahmoud, S. H. (2014). Investigation of rainfall–runoff modeling for egypt by using remote

sensing and GIS integration. Catena, 120, 111-121.

Mahmoud, S. H., Mohammad, F., & Alazba, A. (2013). A GIS-based approach for determination
of potential runoff coefficient for al-baha region, saudi arabia. 2013 International
Conference on Sustainable Environment and Agriculture IPCBEE,

Masoud, M. (2014). Rainfall-runoff modeling of ungauged wadis in arid environments (case

study wadi Rabigh—Saudi arabia). Arabian Journal of Geosciences, 8(5), 2587-2606.

Melesse, A. M., & Shih, S. (2002). Spatially distributed storm runoff depth estimation using
landsat images and GIS. Computers and Electronics in Agriculture, 37(1), 173-183.

Mockus, V. (2010). Estimation of Direct Runoff from Storm Rainfall.National Engineering

Handbook, Section 4– Hydrology,

O'Loughlin, G., Huber, W., & Chocat, B. (1996). Rainfall-runoff processes and modelling.
Journal of Hydraulic Research, 34(6), 733-751.
 Ponce, V. M., & Hawkins, R. H. (1996). Runoff curve number: Has it reached maturity? Journal
of Hydrologic Engineering, 1(1), 11-19.

Ramakrishnan, D., Bandyopadhyay, A., & Kusuma, K. (2009). SCS-CN and GIS-based
approach for identifying potential water harvesting sites in the kali watershed, mahi river
basin, india. Journal of Earth System Science, 118(4), 355-368.

Schultz, G. (1994). Meso-scale modelling of runoff and water balances using remote sensing and
other GIS data. Hydrological Sciences Journal, 39(2), 121-142.

Senay, G., & Verdin, J. (2004). Developing index maps of water-harvest potential in africa.
Downloaded by [University of Exeter] at 02:22 19 July 2016

Applied Engineering in Agriculture, 20(6), 789-799.

Taguas, E., Gómez, J., Denisi, P., & Mateos, L. (2015). Modelling the rainfall-runoff
relationships in a large olive orchard catchment in southern spain. Water Resources
Management, 29(7), 2361-2375.

Vaze, J., Chiew, F., Perraud, J., Viney, N., Post, D., Teng, J., . . . Goswami, M. (2010). Rainfall-
runoff modelling across southeast australia: Datasets, models and results. Australian Journal
of Water Resources, 14(2), 101.

Isaev, V., & Mikhailova, M. (2009). The hydrography, evolution, and hydrological regime of the
mouth area of the Shatt al-Arab River. Water resources, 36(4), 380-395.

Allen, R.G., Perira, Raes D., and Smith, M., 1998. "Crop Evapotranspiration, Guide lines for
Computing Crop Water Requirements", Irrigation and Drainage Paper No. 56, Food and
Agriculture Organization of the United Nations.

Chander G., Markham Brian L., Helder Dennis L., 2009, "Summary of current radiometric
calibration coefficients for Landsat MSS, TM,ETM+, and EO-1 ALI sensors".
International Journal of Remote Sensing of Environment 113 893–903.
 USDA, 2004. Estimation of direct runoff from storm rainfall. National Engineering Hand- book
p. 79.

FAO, 1974. FAO-UNESCO Soil Map of the World, 1:5.000.000. UNESCO, Paris.

Jensen, John R. 1996. "Introductory Digital Image Processing". Prentice Hall, 2nd. Edition, New
Jersey, NY, 750 P.
Downloaded by [University of Exeter] at 02:22 19 July 2016

Jensen, J. R. 2005. "Introductory Digital Image Processing" 2nd Edition, Open Access, Remote
Sensing. ISSN 2072-4292, Prentice Hall.

Janssen, L. (2000). Principles of Remote Sensing, the International Institute for Aerospace
Survey and Earth Science (ITC).
 Captions of Figures:

Figure 1. Location of the study area overlapped on open street map

Figure 2. Overall research activities

Figure 3. Soil texture of the study area derived from USGS soil GIS layer

Figure 4. Land cover map derived from Landsat OLI using support vector machine algorithm

Figure 5. (a) DEM of the study area (b) land slope derived from DEM

Figure 6. Monthly average of rainfall

Downloaded by [University of Exeter] at 02:22 19 July 2016

Figure 7. (a) Curve number, (b) slope adjusted curve number

Figure 8. Monthly average of runoff depth derived from land use and soil information in GIS

Figure 9. Relationship between mean annual rainfall and modeled runoff coefficient
 Table 1: Data sources used to carry out the study

No. Data Source

1 Soil Map FAO website (http://www.fao.org/)
2 Soil Texture Iraq Exploratory Map
3 Satellite Image Landsat OLI (USGS)
4 Rainfall Al-Hila Meteorological Station
5 DEM SRTM (USGS)
Downloaded by [University of Exeter] at 02:22 19 July 2016

6 Topographic Map http://www.diva-gis.org/

 Table 2: Main soil characteristics

Soil Name Sand Loam

Total available soil 60 290

moisture (mm /meter)

Maximum rain infiltration 40 40

rate (mm/meter)

Maximum rooting 900 900

depth(cm)
Downloaded by [University of Exeter] at 02:22 19 July 2016

Initial soil moisture 0 0

depletion (mm/meter)

Initial available soil 60 290

moisture (mm/meter)
 Table 3: average montrhly change in storage and related information for the study area

Month Precipitation (mm) Q (mm) ETa (mm) Ir (mm) S (mm)

Jan 22.5 22.26 61.01 0 -60.77
Feb 16 15.83 64.2 0 -64.03
March 15.6 15.43 64.25 2750 2685.92
April 14.2 14.05 66.39 9856 9789.76
May 3.23 3.19 162.1 18370 18207.94
June 0.31 0.30 257.7 20861 20603.31
Downloaded by [University of Exeter] at 02:22 19 July 2016

July 0.2 0.19 240.6 22957 22716.41

Aug 0.46 0.45 277.1 14230 13952.91
Sep 0.82 0.81 247.7 5962 5714.31
Oct 5.6 5.54 218.2 0 -218.14
Nov 16.1 15.93 188.8 0 -188.63
Dec 18.8 18.60 58.24 0 -58.04
Average 9.49 9.39 158.85 13569.43 7756.74
Total 93080.95
Downloaded by [University of Exeter] at 02:22 19 July 2016
Downloaded by [University of Exeter] at 02:22 19 July 2016
Downloaded by [University of Exeter] at 02:22 19 July 2016
Downloaded by [University of Exeter] at 02:22 19 July 2016
Downloaded by [University of Exeter] at 02:22 19 July 2016
Downloaded by [University of Exeter] at 02:22 19 July 2016
Downloaded by [University of Exeter] at 02:22 19 July 2016
Downloaded by [University of Exeter] at 02:22 19 July 2016
Downloaded by [University of Exeter] at 02:22 19 July 2016