Zoning of Environmental Hazards in Tourism Destinations with Emphasis on Flooding (Case Study: Sarvabad County, Kurdistan Province)

The occurrence of environmental hazards has multidimensional and widespread damages on the natural environment and human society. Among environmental hazards, flooding is one of the most important and frequent hazards, which causes significant human, financial, and environmental losses every year. In different regions, tourist destinations, in addition to the permanent community living in them, welcome a significant number of tourists. Compared to the local community, these tourists are not very familiar with the destination environment and may be at greater risk. Therefore, identifying flood-prone areas in tourist destinations can be an important tool for reducing the damages caused by it. Sarvabad County is one of the most important tourist destinations in Kurdistan Province, due to its natural and cultural attractions. Therefore, the present quantitative and applied research has been conducted with the aim of zoning flood risk in Sarvabad County. In this study, 14 criteria affecting flood occurrence were used to achieve the main objective of the research. Data-driven approach and modern remote sensing tools, geographic information system and MaxEnt model were used to analyze the data. The results of the study showed that due to the mountainous nature of the region, a limited part of Sarvabad County is in high-risk classes in terms of flood risk. In this way, of the total area studied (1044 square kilometers), 96.84 percent is in very low and low risk zones, 1.2 percent is in medium risk zones and 1.96 percent is in high and very high-risk zones. The results also showed that most of the high and very high-risk zones are located in a strip shape in the northern to central parts of the county. The results of the validation of the final research model using the Omission and Predicted Area diagrams and the ROC curve showed that the final model had high accuracy and performance and was able to zone Sarvabad County in terms of flood occurrence.

 
Extended Abstract
 
Introduction
Despite the human, financial, environmental, etc. damages caused by floods, as well as the fact that a large part of our country's geographical space is flooded and the development of unplanned urbanization, studies indicate that flooding, as one of the natural disasters, can be properly monitored using modern technologies and information systems. The first step towards reducing the harmful effects of floods is to identify flood-prone areas and zoning these areas in terms of flood risk, so that, based on the results obtained, principled and optimal decisions can be made about the use of various agricultural, industrial, service land uses and the location of urban and rural settlements, and the harmful effects of floods can be minimized as much as possible. In order to achieve this goal, the use of a geographic information system (GIS) is one of the most important tools for identifying and zoning different areas in terms of flood risk and even other environmental hazards. Sarvabad County, with a population of 43,700, consisting of 8,800 urban and 34,900 rural residents, is one of the border counties located in western Iran and is always at risk of flooding due to its mountainous location, significant rainfall, and the rapid growth of physical development in urban and rural areas. Due to its natural and cultural attractions, this county is located within the Horaman Cultural Landscape, which was registered as a World Heritage Site by UNESCO in 2021. For this reason, it receives a significant number of tourists throughout the year. Therefore, flood risk management in this tourist destination has double value. In order to manage this risk, one of the most important measures is the zoning of Sarvabad County, relying on the geographic information system, so that high-risk spaces can be identified and necessary measures can be taken to take management measures in the future. Therefore, considering the multidimensional damages of flood risk as well as the human and natural conditions prevailing in Sarvabad County, the researchers in the present study are trying to answer the following questions: What is the spatial distribution of flood risk in different risk classes (very high risk to very low risk) in Sarvabad County? How is the validation of the final flood risk zoning model in Sarvabad County?
 
Methodology
This research is among the quantitative and applied researches that have been conducted with the aim of zoning flood risk in Sarvabad County using a data-driven approach and modern remote sensing tools and Geographic Information System (GIS). This study has been written in several systematic stages, considering the overall process governing it. In the first step, the factors and parameters affecting flood occurrence were identified by reviewing scientific records, reliable sources, and expert analyses. Then, these layers were converted to ASCII format for use in the modeling environment to enable their entry into the MaxEnt software. MaxEnt, as one of the powerful models based on maximum entropy, has a high ability in modeling the spatial potential of phenomena based on occurrence points and environmental variables. On the other hand, flood points in the study area were identified through past flood zoning maps, image analysis, and field observations, and after filtering, they were converted to CSV format to be used as training data in the MaxEnt model. In this study, the training and learning data were divided in a ratio of 20 to 80. Also, the number of points extracted was 250 flood points. This data plays a vital role in training the model and determining the relationship between environmental factors and spatial patterns of flooding. After running the model in the MaxEnt environment, a flood probability map was produced for Sarvabad County, which distinguishes different areas in terms of flood risk (very low risk, low risk, medium, high, and very high). The modeling results were also evaluated for validation with statistical indicators such as AUC to ensure the accuracy of the model. In the final step, the modeling outputs were extracted in the GISPro software environment and the generated maps were reviewed and analyzed.
 
Results and Discussion
The results of the study showed that due to the mountainous nature of the region, a limited part of Sarvabad County is in high-risk classes in terms of flood risk. In this way, of the total area studied (1044 square kilometers), 96.84 percent is in very low and low risk zones, 1.2 percent is in medium risk zones and 1.96 percent is in high and very high-risk zones. The results also showed that most of the high and very high-risk zones are located in a strip shape in the northern to central parts of the county. The results of the validation of the final research model using the Omission and Predicted Area diagrams and the ROC curve showed that the final model had high accuracy and performance and was able to zone Sarvabad County in terms of flood occurrence.
 
Conclusion
In line with flood risk management in Sarvabad County, the most important practical suggestions of the present study are: Accurate notification of the exact time of flood occurrence by the Meteorological and Hydrological Organization, distribution of zoning maps among urban and rural managers, installation of warning signs in the vicinity of areas at high and very high risk in terms of flood risk, prevention of construction and development of other physical activities in high-risk areas, accurate determination of hazard zones, especially in the vicinity of residential areas, efficient and optimal multilateral management of pastures and forests, construction of relief centers in high-risk areas, use of flood-resistant structures when developing various types of physical and structural infrastructure, and construction of diversion barriers to direct water to areas further away from residential centers..