Dr. Javad Sadidi, Mrs. Fatemeh Tamnia, Dr. Hani Rezaian,
Volume 8, Issue 4 (1-2021)
Abstract
Nowadays, deep learning as a branch of artificial intelligence acts as an alternative for human with hopeful outcomes. Open Street Map as the biggest open source data is used as a complementary data sources for spatial projects. It is notable that is some advanced counties the accuracy of VGI data is higher than governmental official data. This research aims to use artificial intelligence to produce and subsequently promote completeness of OSM data. Res_UNet architecture was utilized to train landuse categories to the network. The result shows that IoU metric is about 83 percent that implies a high accuracy paradigm. Then, united-based method was used to calculated completeness of OSM data. The unit-based results show that completeness of building blocks, forest, fruits garden and agriculture land are: 3.6, 9.7, 90.4 and 81.88 respectively. It shows the low volunteer participation rate to produce OSM data. On the other side the high accuracy achieved by deep learning leads us to complete OSM data by artificial intelligence instead of human prepared data. The advantage of using machine rather than human may be utilized in undeveloped countries or low density population regions as well as inaccessible areas.
Dr Mohammad Mahdi Hosseinzadeh, Dr Ali Reza Salehipor Milani, Mis Fateme Rezaian Zarandini,
Volume 10, Issue 1 (5-2023)
Abstract
Introduction
A flood is a natural disaster caused by heavy rainfall, which causes casualties and damage to infrastructure and crops. Trend of floods in the world increasing due to climate change, changing rainfall patterns, rising sea levels in the future, and in addition, population growth and urban development and human settlements near river have caused floods to become a threat to humans. One of the most important and necessary tasks in catchments is to prepare flood risk maps and analyze them. In recent decades, researchers have been using remote sensing techniques and geographic information systems to obtain flood risk maps in an area. Due to the numerous floods that have occurred in the Neka river catchment, it is necessary to conduct a study entitled zoning of flood sensitivity in Neka river catchment for more effective management in this area.
Materials and methods
Study area: Neka river catchment area with an area of 1922 Km2 is part of Mazandaran province in terms of political divisions. This basin is between 53º 17´ 54 º44´ east and 36 º 28 ´to 36 º 42´ of north latitude. The highest point of the basin is 3500 m (Shahkuh peak) and the height of the lowest point of the basin in the Ablo station is about 50 m and at the connection to the Caspian Sea is -27 meters. The seven sub-basins of this basin are Laksha, Golord, Burma, Metkazin, Kiasar, Alarez and Sorkh Griyeh. Geologically, the basin is mostly of calcareous and marl formations. In the south and southwest of Neka River, the rock material is mostly clay and calcareous marl, which makes this basin has a high erosion potential
To study the flood zoning of the area using a multi-criteria decision model, 1: 25000 maps of the surveying organization and a digital elevation model with a resolution of 12.5 meters (Alos Palsar) were extracted. In order to study the flood risk in Neka river, 4 criteria of height, distance from the river, land use and slope have been used. In the present study, modeling and preparation of flood risk zoning map in 4 stage including descending valuation, normalization of each class, normalized index weight and integration of criteria has been done by the following linear weighting method. Performing linear weighting operations depends on the weighted average of a number of selected parameters in the opinion of the expert. According to the weight assigned to each criterion based on the expert opinion, each of the criteria was multiplied by the assigned weight and at the end the criteria were added together and the final zoning map was obtained.
Results and Discussion
In this study, using a multi-criteria decision-making system model, a flood risk zoning map in the Neka river catchment was prepared. According to the weight assigned to each criterion based on expert opinion, the final risk probability map has a value between 0.02 to 0.2, which is ultimately divided into 5 classes in terms of flood risk. Value range 0.02 to 0.06 component of very low risk zone, range 0.08 to 0.11 component of low-risk zone, range 0.11 to 0.13 component of medium-risk zone, range 0.13 to 0.16 component of high-risk zone, and finally domain 0.16 to 0.20 components of the area with very high risk potential have been obtained. According to the final divisions in the flood risk zoning map of the catchment area, a safe area means areas where the probability of flooding is very low and close to zero, and in contrast, the area with a high and very high risk potential for flooding has the probability of high-risk floods. According to the final flood risk zoning map, about 982 Km2 (51%) has high and very high vulnerability, as well as about 510 Km2 (26.69%) has medium vulnerability in Neka catchment area.
Conclusion
The results obtained from the model indicates that the highest risk of flooding points are located in the western parts of the Neka catchment area and the end of the catchment area that reach the city of Neka. According to the research findings, the most important factors in increasing the risk of floods were the slope in this area and the distance from the drainage network. According to the results of the model, a large area of the basin is a component of high risk zone, that means the Neka river watershed has a high potential for floods. Evidence and documented reports show that the Neka river Basin has experienced several floods in the last two decades. The major part of the occurrence of floods is due to the natural conditions of the basin, thus it is necessary to reduce flood damage by changing the locations of various land uses based on flood vulnerability maps. Using multi-criteria decision making method can be used to prepare flood risk zoning maps in basins that do not have hydrometric data; It is also a more cost-effective method in terms of time. One of the important issues in the final result of this model is due to the weight of the layers, which should be used by experts, who are familiar with the region and this method and adapt to field evidence.
Keyworlds: Flood, Multi-criteria decision making system(MCDA), Hazard zoning, Nekarod, Natural hazard.