© by PSP Volume 17 – No 11b. 2008 Fresenius Environmental Bulletin
1964
DETERMINATION OF FLOOD RISK IN SAKARYA
SUB-BASIN USING REMOTELY SENSED DATA AND GIS
Orkan Ozcan
1*
, Nebiye Musaoglu
2
, Dursun Zafer Seker
2
and Aysegul Tanık
3
1
ITU, Informatics Institute, Satellite Communication and Remote Sensing Programme, 34469 Maslak, Istanbul/ Turkey
2
ITU, Faculty of Civil Engineering, Geodesy and Photogrammetry Department, 34469 Maslak, Istanbul/Turkey
3
ITU, Faculty of Civil Engineering, Environmental Engineering Department, 34469 Maslak, Istanbul/Turkey
Presented at the 14
th
International Symposium on Environmental Pollution and
its Impact on Life in the Mediterranean Region (MESAEP), Sevilla, Spain, 10 – 14 Oct. 2007
ABSTRACT
Between 1970-2005, 936 floods occurred, 17664393 ha
agricultural land was damaged, and 604 people lost their
lives, in Turkey. The floods have been mostly recorded in
the Sakarya basin, selected as study area. In order to ana-
lyze all kinds of changes in the study area, SPOT XS (20
m) and SPOT PAN (10 m) satellite sensor images within
the years of 1999 and 2006 were used. Land-use/cover
of the region was evaluated from satellite images. Ob-
tained classification results were verified by the field study.
Topography and the drainage qualities of the land were ana-
lyzed by DTM, produced from topographic maps of scale
1:25 000. In addition, meteorological data were collected
and included as inputs to the system. The parameters of
the basin were modeled with the Hydrologic Engineering
Centers River Analysis System (HEC-RAS) software. At
the end of the modeling, areas under flood risk in the study
area and land-use/cover were calculated to be 39.5 km
2
.
When the results were examined, 6.2 km
2
of the risky
areas were evaluated as residential, and the rest as agricul-
tural areas.
KEYWORDS:
Remote Sensing, flood, Sakarya, HEC-RAS.
INTRODUCTION
Floods are among the most serious natural disasters
that engender loss of life and property. It is impossible to
prevent flood disasters only due to meteorological phenom-
ena. Especially in the developing countries like Turkey
being still on the way to economical development, urbani-
zation, triggered by industrialization and sector diversity,
greatly increases the variety and intensity of human activi-
ties in several fragments of river basins. This fact annihi-
lates the hydrological stability in the entire basins leading
to devastating floods, resulting in vast amounts of life and
property losses. The land-use changes observed by the con-
struction of new residential areas, and transportation net-
work extending in the river basins, are important outcomes
of rapid and uncontrolled urbanization, which also drasti-
cally destroys meadows and forests. Such activities, together
with the intensive use of improper agricultural methods,
accelerate intensity and damage of floods.
Natural disasters can nowadays be evaluated and con-
tingency plans and mitigation programs can be prepared,
prior to a new event by means of Remote Sensing (RS) tech-
nology utilizing satellite images via extracting necessary data
and executing necessary risk analysis. RS provides a reli-
able and cost-effective way for field data collection, allow-
ing for continuous and large-area coverage of many vari-
ables.
For the last two decades, advancement in the field of
RS and Geographic Information System (GIS) have greatly
facilitated the operation of flood mapping and flood risk
assessment [1]. The main advantage of using GIS for flood
management is that it not only generates a visualization of
flooding, but also creates potential to further analyze this
product to estimate probable damage due to floods [2, 3].
The aim of this study is to evaluate the flood risk analysis
(an important issue) in Sakarya sub-basin using RS and
GIS. In the study area, a hydrological model is developed on
the possible areas under flood risk, and related environ-
mental effects were analyzed.
THE STUDY AREA
Sakarya is one of the 81 provinces of Turkey located
in the Marmara region. Between 1970-2005, 114 floods
occurred in this basin, 8 644 ha of agricultural land was