Page 279 - NGTU_paper_withoutVideo

Modern Geomatics Technologies and Applications

Flood Risk Mapping Using by GIS-based Multi-Criteria Decision-Making
(A Case Study: Miandoab Basin)

1
Mohammad Sadegh Tahmouresi , Mohammad Hossein Niksokhan , Iman Zandi , Mohammad Hasan
1
1*
1
Goudarzi

1 Department of Environmental Eng., School of Environment, College of Eng., University of Tehran
2 Department of GIS, School of Surveying and Geospatial Eng., College of Eng., University of Tehran
Corresponding Author: niksokhan@ut.ac.ir

Abstract: In recent years, floods have affected Iran and caused significant damage to various sectors. Having the right
information on flood-prone areas helps prevent, control and manage this crisis. In the present study, a combination of a
Geographic Information System (GIS), Satellite Imagery, and Fuzzy Analytical Hierarchical Process (AHP) has been used
to zoning flood risk in the Miandoab basin. The weighting method results showed that Proximity to Rivers and Rainfall
are the most important criteria in flood risk zoning, respectively. The results of flood risk zoning showed that the
southwestern areas have a high potential for flooding.

KEYWORDS: Flood Risk Mapping, Miandoab, GIS, Fuzzy AHP.

1. Introduction

One of the natural and catastrophic dangers is floods, which cause significant damage to people's lives and property [1]. Floods
result from significant rainfall and snowmelt that creates an overflow situation in the river and temporarily stagnates in the lands
along the river [2]. Floods are a complex phenomenon, which is why researchers are conducting extensive research in diverse
areas to manage, control, and prevent potential damage [3]. There are several natural and human factors involved, which can
lead to a catastrophic flood. Recently, climate change has become an important factor in severe flood events[3].

Charlton et al. [4] argued that flood risk in an area could be significantly affected by climate change, and it possibly can change
land-use patterns. Furthermore, it creates an impenetrable surface, which may increase the flow rate. In addition to climate, the
various factors that depend on the occurrence of floods such as elevation characteristics of an area, slope, proximity to the main
rivers, soil texture, topographic curvature [5]. According to Opolot [6], between 2000 and 2008, about 100 million people were
affected by severe floods worldwide. These events are more concentrated in developing countries, where urbanization and
civilization along rivers are growing rapidly. Deforestation to establish settlements and river seizure are considered flood
accelerator factors [7].
From the last two decades, various methods including hierarchical analysis process [5], fuzzy logic and genetic algorithm [5],
fuzzy theory [8], a hydrological forecasting system, a decision tree model, Multi-Criteria Decision Making (MCDM) and spatial
techniques have been developed to study flood risk. In recent years, Geomatics, Geospatial Information System (GIS), and
Remote Sensing and their integration with MCDM have been widely used in crisis management. Chakraborty and
Mukhopadhyay [9] proposed a combination of Analytical Hierarchical Process (AHP) and GIS to flood zoning in India.
Pourghasemi et al. [10] Used a combination of machine learning methods and satellite imagery to flood risk mapping. Panahi et
al. [11] have used integrating an improved analytical network process with statistical models to the flood zoning process in Iran.

2. Methodology
In the present study, to determine flood-prone areas in the Miandoab basin, a combination of GIS, satellite images, and Multi-
Criteria Decision Analysis (MCDA) has been used. For this purpose, by studying previous research and surveys of experts,
appropriate criteria for determining flood-prone areas were discovered. Then, by using GIS technology and Remote Sensing,
decision criteria maps were prepared. In the next step, the weights of the decision criteria were calculated by the fuzzy AHP and
matrix of pairwise comparisons completed by experts. Finally, by considering criteria weights and criteria maps prepared by

1