Polytechnic Journal. 2022.

12(2) 129-135
ISSN: 2313-5727
http://journals.epu.edu.iq/index.php/polytechnic

R ES E A R CH AR T I C L E

Preparation of Curve Number Grid Map of Rawanduz River

Basin in KRI Using GIS Techniques
Israa D. Ahmed1*, Jehan M. Fattah Sheikh Suleimany
1
Department of Water Resource Engineering, College of Engineering, Salahaddin University, Erbil, Kurdistan Region, Iraq
2
Department of Water Resource Engineering, College of Engineering, Salahaddin University, Erbil, Kurdistan Region, Iraq

*Corresponding author: A B S T R AC T
Israa D. Ahmed,
Department of Water
Resource Engineering, The study is being conducted with the goal of applying various models and techniques for catchment
College of Engineering, delineation and providing the opportunity to determine the Curve Numbers (CN) for the Rawanduz
Salahaddin University,
river basin (2956 Km2), which is located in the northeastern section of Iraq, Kurdistan region, by
Erbil, Kurdistan Region,
Iraq. utilizing Geographic Information System (GIS). Satellite data (Landsat 8 OLI) for 2020 was utilized to
build land cover classes, with a classification accuracy of 90% and a Kappa index of 86.18%. The
E-mail:
hydrologic soil map for the region is digitized from the Iraqi soil map. In specifically, Soil Conservation
israa.ahmad@su.edu Service-Curve Number (SCS-CN) is technique that will be utilized to predict direct surface runoff. This
.krd technique incorporates a range of significant watershed features, such as Hydrological Soil Groups
(HSG) and land use/land cover (LU/LC), to be utilized as input variables for numerous hydrologic
Received: 6 May 2022 processes. In terms of land use and hydrologic soil group combination, the lowest CN value was found
Accepted: 4 October 2022
to be 58 in forest areas, which corresponds to an increased ability of the soil to retain rainfall and will
Published: 1 February. 2023
produce much less runoff, and the highest CN value was found to be 92 in medium-residential areas,
DOI which causes most of the rainfall to appear as runoff, with minimal losses.
10.25156/ptj.v12n2y2022.pp12
9-135

Key Words: GIS, Soil map, Land use/land cover, curve number, Rawnduz river
basin.

1986). Modeling is frequently utilized to decide CN.

1-INTRODUCTION ESRI (Environmental Systems Research Institute)
The Curve Number is utilized in hydrology to calculate created ArcMap GIS for evaluating, constructing,
how much more rainfall is swelling through land or modifying, retrieving, and displaying different data
underground and also how much water generates surface inputs (Wang and Cui, 2004). The CN approach has been
water runoff. A CN is an index created by the Soil incorporated into a few hydrological processes, like
Conservation Service (SCS) and now known as the Storm Water Management Model (SWMM) (James et
Natural Resource Conservation Service (NRCS), is al., 2010), SWAT (Soil and Water Assessment Tool)
utilized to measure the amount of rainfall which (Arnold et al., 1996), WMS (Watershed Modeling
infiltrates the soil as well as the amount of surface runoff System), Hydrological Modeling System (HEC-HMS)
(Hawkins et al., 2008). Since of the easy, flexibility, and utilizes this approach to calculate surface runoff
accessibility for ungauged basins, this strategy has been (Shadeed and Almasri, 2010). The CN is critical in either
very well known inside the US (United States) and such hydrologic models, like the rainfall-runoff modeling, for
other nations (Sartori et al., 2011). It was, to begin with, determining maximum discharges and runoff volume.
created for agricultural basins, and after that it was thus
Manually calculating CN for large areas or multiple
utilized in the region of urban areas. This approach
drainage watersheds can be tedious and time-consuming.
should only be utilized with twenty - four - hour rainfall
Thus, GIS (Geographical Information System) is an
periods (Durrans et al., 2003, USDA, 1986). As it was
excellent tool for such a project. HEC-GeoHMS is a
not affected by time, it ignores variances produced by
geospatial hydrological modeling tool created with HEC
variable rainfall lengths and concentrations (USDA,
Polytechnic Journal ● Vol 12 ● No 2 ● 2022 129 ©2011-2022, Erbil Polytechnic University, Erbil, Kurdistan Region - F.R. Iraq
of rapid creation for hydrologic processes instead of DEM utilized in this investigation was taken from
traditional approaches (Manual, 2003). The use of Website of the US Geological Survey (USGS). The
remote sensing (RS) and GIS is dependable way for DEM is used to derive the slope, basins, gradient and
preparing majority of the incoming data needed by the drainage pattern. To begin geometric rectification, the
SCS-CN equation. It was created for the ArcMap system DEM (30m) spatial resolution was projected to the
and uses the Spatial Analysis extension. The CN of a Universal Transverse Mercator (UTM) data processing
river basin is determined by utilizing a combination of systems of Datum WGS 1984 (Zone-38). Rawanduz
land use, type of soil, and DEM (Digital Elevation river basin DEM is illustrated in Figure 2.
Model) datasets. Hydrologic soils are divided into 4
types: A, B, C, and D (USDA, 1986). Group of A has a
large capacity of infiltration, whereas Group of D has a 2- Land Use
lower capacity of infiltration.
The Landsat image (Landsat 8 OLI) used in this
Several researches appear to agree that utilizing investigation was obtained from the United States
GIS to calculate CN is reliable and effective utilize Geological Survey (USGS) Earth Explorer site
of time such as .(Topno et al., 2015, Gajbhiye and (https://earthexplorer.usgs.gov/). LULC have a
Mishra, 2012) using the SCS-CN equation significant impact on infiltration and the amount of
integrated with GIS for runoff prediction. This is a runoff. For the classification of images to be effective,
popular hydrological model for calculating runoff several variables should be considered, counting the
based on rainfall and CN. Remote sensing and GIS, accessibility of high-quality images of Landsat, an
combined with the SCS-CN model, are shown to be accurate classification technique, and the users'
an effective tool for runoff prediction.(Zhan and competence and experience with processes. (Rwanga and
Ndambuki, 2017). The analysis and classification
Huang, 2004) utilized the tool of ArcCN Runoff, an
method of different LULC classes were carried out using
application of the ESRI ArcGIS program, to of two Landsat image 8 OLI (Operational Land Imager)
compute runoff and generate CNs during a rainfall images of satellite were obtained in 2020. The
events inside a basin. The major purpose of this Supervised Classification Technique was done utilizing
research is to create the CN grid map to find the the Maximum Likelihood Algorithm in ArcGIS.
runoff of the watershed. This study contains Supervised classification concurring to (Eastman, 2003)
obtaining the spatial dataset, land use/land cover, occurs when "the user creates the spectral signatures of
type of soil, and CN grid using Hec-GeoHMS common classes, like urban and forest, and then the
extension in the ArcMap program assigns each pixel in the picture to the land
cover to which its signature is most similar." Supervised
classification was used after defining an area of study,
2-METHODS AND MATERIALS which is referred to as training classes. To represent a
certain class, more than one training area was employed.
2-1 Study Area The training locations were chosen in accordance with
The watershed is situated in the northeastern section of the Landsat Image, Google Earth, and Google Maps. The
Iraq, in the Kurdistan region, near Erbil, located between main supervised classification operation sequence was as
latitude (36°23´ and 36°80´ North) and longitude (44°16´ follows: defining training sites, extracting signatures, and
and 45°10´ East) and has an area of 2956 km^2. The classifying the Landsat image. The Maximum
Rawanduz river watershed goes through the villages of Likelihood Algorithm in ArcGIS is the most commonly
Rayat and Choman. Based on the tectonic map (Buday used for image classification.
and Jassim, 1984), It is located inside the High Folded
Zone, Imbricated Zone, Unstable Shelf, and Zagros 3-The Accuracy Assessment Estimator and the
Suture Zone. Figure 1 illustrates a map of study area Kappa coefficient for Image classification
The Kappa index and accuracy assessment is the most
critical and is the final step in the image classification
2-2 Processing of Data process (Foody, 2002). The terms "accuracy evaluation"
ArcGIS ver. 10.8, Arc Hydro Tools and Hec-GeoHMS and "Kappa index" refer to the comparison of the
were utilized to pre-process the obtained data for the classified image to another data source that is thought to
watershed. be accurate or ground truth data. Accuracy makes the
value of information of the resulted data to the user. This
1- Digital Elevation Model (DEM) procedure employs three geospatial tools in ArcMap:

Polytechnic Journal ● Vol 12 ● No 2 ● 2022 130

Construct Accuracy Assessment Points, Modify Matrix (Arumugam et al., 2021).
Accuracy Assessment Points, and Calculate Confusion

Figure 1: Location map of Rawanduz river ungauged basin map

4- Soil map preparation

The map of soil of the Kurdistan area of Iraq was
received from the Agriculture Department and Water
Resources of the Kurdistan Region Administration, Iraq.
The Iraq soil map was downloaded from The Global
Hydrologic Soil Groups (HYSOGs250m) for the Curve
Number-Based Runoff soil map. Internet sites in 2018
and updates in 2020 (Ross et al., 2018). The soil map
was geo-referenced using ArcGIS, and then the soil data
for the watershed was extracted by using ArcGIS tools.
Figure 3. illustrates the soil map of the Rawanduz river
basin, which includes four distinct types of soil: clay
loam, loam, sandy clay, and sandy clay loam. Therefore,
the hydrologic soil group (HSG) is the information
needed to compute a CN. There are four groups (A, B, C,
and D) and three dual groups (A/D, B/D, and C/D).
Figure 2: Digital elevation model map of Rawanduz
Table 1 defines each of these terms.
river basin (https://earthexplorer.usgs.gov/).

Table 1: Illustrates the HSG and the associated soil texture (Maidment, 2011)

HSG Soil Type Texture Runoff Potential Rates of Infiltration

A Sand, loamy sand and sandy loam Low High

B Silty loam and loam Moderate Medium

Polytechnic Journal ● Vol 12 ● No 2 ● 2022 131

 C Sandy clay loam High/moderate Low

D Silty clay loam, sandy clay, silty clay, and clay High Very Low

3-RESULT AND DISCUSION

The map of LU/LC and HSG map have been generated
for development of CN grid map, and the findings are
given below.

3-1 Hydrological Soil Group (HSG) Map

In ArcGIS, the hydrologic soil group was computed and
used as an input data for the CN grid map. The soil code
has been allocated to each soil group in the basin. The
HSG of B, C, C/D, and D groups are shaped in ArcMap
software as illustrates in Figure 5.
Figure 3: Soil Map of the Rawanduz Basin with soil PctA, PctB, PctC, and PctD are the names of four new
texture (HYSOGs250m) fields. For Rawanduz river basins, only one soil group is
assigned to each polygon, so a polygon with soil group
"A" will have PctA = 100, PctB = 0, PctC = 0, and PctD
5-Curve Number (CN) Grid map
= 0. Likewise, for a polygon with soil group D, only
The data for land use map, map of soil, and map of DEM PctD = 100, while the other three Pcts are zero. For a
were used in GIS ArcMap. The CN grid map was created polygon with soil group C/D, it will have PctC = 50 and
utilizing ArcGIS's HEC-GeoHMS extension tool. The PctD = 50, and the other two Pcts are zero. The attribute
maps of soil and land use of the study region were table shown below was produced. The Soil Code will be
integrated utilizing Union tool inside the Arc Toolbox, populated with the letters A, B, C, and D, as illustrated in
and a CN Look-Up tables was created utilizing create Figure 6.
table inside Arc Toolbox with each soil land map was
provided. A CN grid map was produced by HEC-
GeoHMS after merging the LULC and HSG maps 3-2 Classes of Land Use/Land Cover
utilizing ArcMap. HEC-GeoHMS utilizes the combined The supervised classification and maximum likelihood
feature class (union) and the lookup table (CN_LookUp) methods are used to classify the images in ArcGIS. The
as an input to create a CN grid map. The conceptual region from each class was computed by dividing the
structure for CN grid mapping is depicted in Figure 4. count of pixel and whole region into consideration

Landsat 8 OLI, raster image in

Land Use map, Supervised
2020 from
classification
(https://earthexplorer.usgs.gov/)

Data Soil map geo-referencing and

Collection extracted study area in HSG map Union of LULC
2020from(HYSOGs250m) and HSG map

DEM raster data of 30m resolution

in 2014 from Fill the DEM of
CN grid map
(https://earthexplorer.usgs.gov/) watershed
Polytechnic Journal ● Vol 12 ● No 2 ● 2022 132
Figure 4: Curve Number (CN) Mapping Conceptual Framework.

(watershed). As a result, the percentage distributions of

Table 2: The area of LULC in the Rawanduz river
each classed region is recorded within Table 2. Rate of
ranges as classified are: Forest (88.4%), Agriculture
(8%), Medium residential (3.5%), and Water body Number Land Use Type Area in Percent
(0.1%). The total classification accuracy was 90%, while
the Kappa coefficient was 86.18%. Figure 7. depicts the 1 Water Body 0.1
several kinds of LU/LC in watershed.
2 Forest 88.4

3 Agriculture 8

4 Medium Residential 3.5

basin

3-3 Curve Number (CN) Grid Map

Figure 8 shows a union attribute table for soil and
LU/LC. Figure 9 shows a lookup CN attribute table. The
right CN were allocated to all combinations using usual
SCS-CN database. The CN parameter is unit less and
Figure 5: The map of HSG of Rawanduz river basin ranges 0 (highest range of infiltration) to 100 (zero rate
(HYSOGs250m) of infiltration). The determined CN with each polygon
would be for average conditions. The CN map was
created by intersecting the Land Use and soil

Polytechnic Journal ● Vol 12 ● No 2 ● 2022 133

Figure 6: Attribute table of soil, soil_group filled populated with letters A, B, C, and D
hydrological group maps. The values of CN in the Figure 9: CN-LOOKUP attribute table
Rawnduz Basin range between 100 and 58. The
weighted CN from the basin is equal to 74. The CN grid
map is illustrated in Figure 10.

Figure 10: Curve Number grid map of Rawanduz

River Basin
Figure 7: LULC of Rawanduz River Basin

CONCULSION

The following findings are described in the current

study:
1- The ArcGIS ver. 10.8 software and the Hec-GeoHMS
extension are appropriate techniques for computing CN.
ArcMap is used to create a watershed delineation, image
classification, merge the LULC and HSG maps, and a
CN lookup table. The CN grid map was produced by the
Hec-geoHMS extension tool using the combined feature
class (union) and the lookup table (CNLookup). The CN
is accurately and simply obtained using the capability of
the CN technique to integrate layers of soil and layers of
land use, which simplifies computations and produces an
accurate CN. The calculated CN without the use of
Figure 8: Attribute tables of the soil and land ArcGIS could be time-consuming.
use/land cover union map
2- In this research, four HSG were recognized: soil types
B, C, C/D, and D. Four different types of LU/LC have
been recognized: these include water, medium-
residential, forest and agriculture. This data is extremely
valuable to modeling of rainfall-runoff in order to predict
runoff parameters and maximum discharge utilizing the
SCS-CN approach.
3- According to the findings, a CN of 100 denotes the
water surface. The maximum CN of 92 indicates high
runoff and low rates of infiltration in medium residential,
whereas the minimum CN of 58 indicates low runoff and
high rates of infiltration in forested area. The CN of 75.5

Polytechnic Journal ● Vol 12 ● No 2 ● 2022 134

refers to forest areas for the HSG as C/D, and the CN of MAIDMENT, D. R. 2011. HandBook of Hydrology,
89.5 indicates the agriculture areas. University of Texas at Astin, Texas USA 1993. Marek,
Mark A., Hydraulic Design Manual, 32611-0570.

REFRENCES MANUAL, U. U. S. 2003. Geospatial Hydrologic

Modelling Extension, HEC-GeoHMS, Version 1.1. US
Army Corps of Engineers Hydrologic Engineering
ARNOLD, J., WILLIAMS, J., SRINIVASAN, R., Center, California, USA.
KING, K. & GRIGGS, R. 1996. SWAT: soil and water
assessment tool. USDA-ARS, Grassland. Soil and Water ROSS, C., PRIHODKO, L., ANCHANG, J., KUMAR,
Research Laboratory, Temple, TX, 53, 1423-1431. S., JI, W. & HANAN, N. 2018. Global hydrologic soil
groups (HYSOGs250m) for curve number-based runoff
ARUMUGAM, T., YADAV, R. L. & modeling. ORNL DAAC.
KINATTINKARA, S. 2021. Assessment and Predicting
of LULC by Kappa Analysis and CA Markov model RWANGA, S. S. & NDAMBUKI, J. M. 2017. Accuracy
using RS and GIS Techniques in Udham Singh Nagar assessment of land use/land cover classification using
District, India. remote sensing and GIS. International Journal of
Geosciences, 8, 611.
BUDAY, T. & JASSIM, S. 1984. Geological Map of
Iraq, 1: 1,000,000 Scale Series. Sheet No. 1, Tectonic SARTORI, A., HAWKINS, R. H. & GENOVEZ, A. M.
Map. Geo. Surv. Min., Inves., Baghdad, Iraq. 2011. Reference curve numbers and behavior for
sugarcane on highly weathered tropical soils. Journal of
DURRANS, S., DIETRICH, K. & AHMAD, M. 2003. irrigation and drainage engineering, 137, 705-711.
Stormwater conveyance modeling and design, Haestad
press. SHADEED, S. & ALMASRI, M. 2010. Application of
GIS-based SCS-CN method in West Bank catchments,
EASTMAN, J. 2003. Idrisi Kilimanjaro. Guide to GIS Palestine. Water Science and Engineering, 3, 1-13.
and image processing, 1-328.
TOPNO, A., SINGH, A. & VAISHYA, R. 2015. SCS-
FOODY, G. M. 2002. Status of land cover classification CN Runoff estimation for Vindhyachal region using
accuracy assessment. Remote sensing of environment, remote sensing and GIS. International Journal of
80, 185-201. Advanced Remote Sensing and GIS, 4, 1214-1223.
GAJBHIYE, S. & MISHRA, S. Application of NRSC- USDA, S. 1986. Urban hydrology for small watersheds.
SCS curve number model in runoff estimation using RS Technical release, 55, 2-6.
& GIS. IEEE-International conference on advances in
engineering, science and management (ICAESM-2012), WANG, X. & CUI, P. 2004. Linkage of ArcView GIS
2012. IEEE, 346-352. with the RZWQM. J Spatial Hydrol, 4, 1-15.
HAWKINS, R. H., WARD, T. J., WOODWARD, D. E. ZHAN, X. & HUANG, M.-L. 2004. ArcCN-Runoff: an
& VAN MULLEM, J. A. Curve number hydrology: ArcGIS tool for generating curve number and runoff
State of the practice. 2008. American Society of Civil maps. Environmental modelling & software, 19, 875-
Engineers. 879.

JAMES, W., ROSSMAN, L. & JAMES, W. 2010.

User’s Guide to SWMM5: Computational Hydraulic
International (CHI). USA.

Polytechnic Journal ● Vol 12 ● No 2 ● 2022 135