شناسایی نواحی مستعد سیلاب در استان گلستان با استفاده از تصاویر سنتینل و مدلسازی هیدرولیکی

Introduction

A flood occurs when the river flow can no longer be contained within its bed, and over spills its banks. Flooding is a natural and regular reality for many rivers, caused by any pulse of overflowing water that overwhelms a river channel, which supports the most naturally dynamic ecosystems. However, humans often perceive floods negatively due to damage and loss of life. Flooding is the most widespread hydrological hazard worldwide that affects water management, nature protection, economic activities, hydromorphological alterations on ecosystem services, and human health. The mitigation of the risks associated with flooding requires a certain management of flood zones, sustained by data and information about the events with the help of flood maps with sufficient temporal and spatial resolution. This paper presents the potential use of the Sentinel-1 images for flood mapping in monitoring the flood that occurred during March 2019 in the Golestan province. More specifically, in this study, we describe accurate and robust processing that allows real-time flood extension maps to be obtained, which is essential for risk mitigation.

The aim of this paper was the detection of area susceptible to flood in the Golestan province using Sentinel-1 SAR images and modeling response of area susceptible to changes of Gorganroud river discharge using the HEC-RAS hydrodynamic model during different return periods. The peak discharge data recorded at Golestan province hydrometer station was employed to predict peak discharge values during different return periods.





Methodology

The Gorgan Plain, northern Iran, is part of the larger Aralo-Caspian depression which experienced strong subsidence during the Middle Pleistocene, resulting in the gentle inclination of the plain to the west (average slope of ~0.6%). In this study, it is been used Sentinel-1 SAR images because of their features, configuration, and the free data set available online from Earth Observation data resources. Sentinel-1 SAR datasets, coming before and after the event, were downloaded via the Copernicus Open Access Hub platform. Also, for the perpose of modeling, the peak discharge data recorded at Golestan province hydrometer station was employed to predict peak discharge values during different return periods.



Results and Discussion

SAR data is preferred for flood mapping and real-time monitoring in all weather conditions. In this study, dual-polarized (VV and VH) Sentinel-1 SAR images coupled with hydrological data (peak discharge data) were used to produce flood inundation maps. Thresholding technique has been applied to determine the flood mapping through Sentinel-1 data. VH and VV polarisation methods have been applied for a comparison of their respective accuracies in delineating surface water. The finding reveals that the most accumulation of flood took place in the channels with a massive semiment accumulation surrounding the agricultural land and residential region. The proposed approach demonstrates that the microwave remote sensing data along with GIS can be used efficiently for flood inundation mapping, monitoring, and analysing its effect on channel morphology. Therefore, the results of this study will help to take the initiative to reduce the flood hazard impact in the doab area and increase the flexibility in the process of flood management.



Conclusion

The present study exhibited area susceptible to flood in the Golestan province using Sentinel-1 SAR images and modeling response of area susceptible to changes of Gorganroud river discharge using the HEC-RAS hydrodynamic model during different return periods. Flood occurs due to rapid population growth, land degradation, and climate change, and causes harmful damages to human beings and properties. This can be minimized by giving attention to flood risk measures. This study was aimed to map flood inundation areas along the Gorganroud River using Sentinel SAR images, GIS, and HEC_RAS. Flood inundation mapping is used to define the zones which are more susceptible to flood along the Gorganroud river. Using the past peak discharge data and the release of floods related to 2019, Golestan, besides topographic data, maps were made to illustrate areas predictable to be covered with the flood for different releases. The flooded areas on Gorganroud have been modeled using peak flows for different reoccurrence eras using the HEC-RAS model, GIS for spatial data handling, and HEC-GeoRAS for interfacing among HEC-RAS and GIS. These critical floods were damaging the areas around the River, which is hazardous to social and economic growth due to loss of lives and destruction of properties. Residential areas and agricultural lands are located along the river banks and are highly susceptible to flooding for all return periods. Generally, this study discovered that flooded areas in the upstream and middle parts of the River are high as related to the downstream parts.

Keywords: Flood management, Environmental planning, HEC-RAS, Sentinel-1.