INVESTIGATION OF RAINFALL- RUNOFF MODELLING OF THE VAN LAKE 
CATCHMENT BY USING REMOTE SENSING AND GIS INTEGRATION 
M.Coskun, N. Musaoglu 
“ ITU, Civil Engineering Faculty, 34469 Maslak Istanbul, Turkey Istanbul 
coskun@be.itu.edu.tr 
KEY WORDS: Remote Sensing, GIS , Landsat TM, Runoff depth, SCS Curve Number , DEM 
ABSTRACT: 
The aim of this study is to determine runoff depth of the Van Lake basin by using remote sensing and geographic information 
system (GIS) integration. While determining the results we used various data sets such as Landsat satellite image, 1:25000 standard 
topographic map, and soil map data. Standard 1:25000 topographic maps were used to derive digital elevation model. Sub-basin area 
and physical characteristics of the study area such as slope and aspect maps were determined with the help of DEM into GIS as 
layers. Digital Elevation Model was used to determine basin model through Hec-GeoHMS . Landsat image was classified by using 
digital image techniques and integrated into GIS with hydrological soil map. Soil Conservation Curve Number method was used to 
determine curve numbers and runoff depth distribution of the basin area. We show that remote sensing and GIS technology are 
suitable for analysis of the runoff depth distribution of the basin area. The proposed method can be applied to predict for ungaged 
watersheds, flood, and other water resource applications. 
1. INTRODUCTION 
For the last years, engineers and planners have been working on 
the modeling of environmental system. An accurate modeling 
of basin will require determination of the spatial and temporal 
distribution of hydrological parameters. Remote Sensing and 
Geographic Information System with increasing the 
advancement of the computer technology have been applying to 
extract land surface properties at spatial and temporal scales 
which are very useful input data for hydrological model. 
Land use and land cover have several impacts 
on the hydrological cycle such as floods, droughts, runoff, 
water- quality. Rainfall-Runoff model play an important role to 
understand hydrological condition of basin areas and predict 
their behavior over time. Accurate process for prediction runoff 
volumes is used to flood warning, navigation, water quality 
management and many water resource applications. Land 
surface characteristics are important to generate rainfall-runoff 
model, distributed model need remote sensing and GIS 
technologies to produce more accurate spatial data which play 
an important role to derive input data such as land use/cover, 
soil type. 
Conventional hydrological model, to estimate 
runoff model input parameters have to be determined through 
ground truth measurement which still need huge economic and 
time-labor consuming. Therefore, remote sensing can also 
provide information about runoff input data most cost-effective 
and large-land coverage. Moreover, these data is suitable to 
enter into GIS. All kinds of data can be stored as different layer 
by using GIS techniques. 
Remote sensing can be used to obtain extremely valuable input 
data for distributed hydrological model. Remote sensing can 
provide measurements of many of the hydrological variables 
2 
0 
8 
used in hydrologic and environmental model applications, 
either as direct measurements comparable to traditional forms, 
as surrogates of traditional forms, or as entirely new data set. 
A.M. Melesse, S.F. Shih ,2002). 
GIS is a computer based system used capture, store, analyze, 
update and display data. GIS handle mainly two classes of data 
which are spatial data, identifying the location and topography 
of map and attribute data, identifying the characteristics of these 
feature. GIS technology provides a flexible environment for 
entering, analyzing, and displaying digital data from various 
sources necessary for urban feature identification, change 
detection, and database development. (Q.Weng,2001). With the 
aid of a GIS, distributed parameter of hydrological model, 
digital elevation model (DEM) processing have become popular 
analysis watersheds topographic. 
In this study, Soil conservation curve number method was used 
to determine runoff depth with the aid of remote sensing and 
GIS technologies. 
2. DESCRIPTION OF STUDY AREA 
Van Lake locate in the eastern of Turkey at 
about 43°E and 38.5°N, and it is the largest lake in Turkey. Its 
average clevation is 1650 m above mean sea level. The average 
extent of the lake is 3600 km ? It is surrounded by relatively 
large drainage basin of 12 500 km 2 (M Kadioëlu et al, 1997). 
Only in the north-east, near Ercis and Muradiye, and in the 
north-west near Ahlat is a total of 19,000 ha of land irrigated 
with water from streams. The lake has no outlet and is fed by 
numerous streams including the Karasu, Hosap, Güzelsu, 
Bendimahi, Zilan and Yenikôprü Streams. Three of the rivers 
  
Internatior 
Fen 
only have 
parameters 
river inclu 
value. In tl 
about 2 m 
the shore 
governmer 
people's pi 
subdivided 
characteriz 
basin area 
remote sen 
method. (F 
  
Data Sourc 
In the stud: 
elevation n 
maps and v 
have been 
studies that 
May 1987 
TM data ha 
was not ust 
or radiant t 
TM bands 
wavelength 
respectively 
of 0.76-0.9 
with | freq 
respectively 
with 256 le 
from 1: 25 
Digital soi 
(General I 
1:25000-sc: 
KHGM inc 
the map. 
Geomet 
Remote ser 
displacemei 
variations i 
[he intent