عنوان مقاله [English]
Karst zones are spread in large parts of the earth and karst groundwater provides a very important part of the water needs of human societies. One of the manifestations of karst development is the presence of sinkholes in the area. These forms play a major role in infiltrating the volume of many surface runoff and feeding aquifers, and on the other hand, increase the biological risks and the possibility of contamination of aquifer resources. Therefore, identifying sinkholes and preparing their geographical database is of special importance for water resources management. In order to identify the sinkholes in the area, clustering data mining methods (J48 algorithm) along with Geographic Object-Based Image Analysis (GEOBIA) were used. The data used in these methods were data layers of DEM, slope, aspect, flow direction, hillshade, flow accumulation, curvature, profile and plan curvature, which derived from the TanDEM-X digital elevation model with a 12 meter resolution. The results showed that flow accumulation with 45.6%, Max-diff index with 30.2%, slope with 14.8% and aspect with 9% were important in the detecting of syncholes, respectively. Using a combined data mining and GEOBIA approach with TanDEM-X data can provide acceptable results with a synchol detection accuracy of 71.3%. This method is suitable for detecting sinkholes and for detecting areas close to aquifers in large karst massifs.
Karst zones are spread in different parts of the world and according to Ford and Williams (2007), about 20% of the continents' surface is covered with karst, which accommodates 1.4 of the world's population. In addition to its benefits for human beings, karst can also be dangerous if it is not accurately identified. Karst hazards include subsidence and damage to urban and rural facilities, water escape from dam lakes, disruption of road construction activities and water tunnels, as well as the high potential for karst water resources pollution. The latter case, the high potential for water pollution in karst areas, is one of the most important hazards that affects not only the individual or individuals but the entire human community and their health in related areas (Ford and Williams, 1992).
The study area is a limestone massif and part of the Zagros Rorandeh in western Iran, which is located in Kermanshah province and north of Kermanshah and is known as Biston-Parav rugged. This mountain massif with latitude coordinates 34° 21´ to 34° 44´ north and longitudes 46° 53´ to 47° 27´ east is surrounded on all sides by open plains and its area is equal to 1033 square kilometers. Its maximum height is Parav mountain with 3385 meters height.
In the present study, different data and methods were used to form a sinkholes extraction process by shaping. The salient components of this process are described in detail below. The present study is based on experimental method and its data are taken from field visits, library studies and use of GIS and remote sensing systems. First, to get acquainted with the karst literature and the dangers of sinkholes, the research background was studied. Then, through field visits and identification and study of sinkholes and using cameras, simple and laser meters and GPS in areas where it is accessible, sinkholes are identified and the morphometry of sinkholes (number, large diameter, small diameter, depth) is evaluated. it placed. Then, TanDEM-X data with a resolution of 12 meters were taken from the German Space Agency (https://tandemx-science.dlr.de/) and the steps of pre-processing and extraction of the required information layers were performed. Next, the objects were identified using data mining, and the necessary rules for identifying sinkholes were extracted from the information layers and their properties. sinkholes were extracted by applying rules to the data set. Finally, the accuracy of classification was assessed.
Results and Discussion
After applying the data mining process to the samples, the result is a list of object properties that will be used to identify sinkhole and construct a classification tree. The results of classification accuracy in the data mining stage were 87.6%, which seems acceptable. According to the results, the characteristics of cumulative flow with 45.6%, Max-diff index with 30.2%, slope with 14.8% and finally slope direction with 9%, respectively, can play the most important role in detecting sinkhole. By applying the result of the data mining process in the form of a set of rules on the set of images, synchronous identification was performed. The overall accuracy obtained at this stage was equal to 71.3%.
The aim of this study was to extract and identify sinkholes using remote sensing data for Parav-Biston massif located in Kermanshah. The advantage of this study is the use of TanDEM-X data and the combination of data mining and GEOBIA methods, which increases the accuracy and better identification of the desired phenomena. Based on data mining methods and object-oriented classification, the map of sinkholes was produced. Our results show that the sinkhole detection approach using TanDEM-X data provides an effective way to analyze sinkholes in large or inaccessible areas. Which may be very difficult and time consuming on land. Using this approach, considering the relationship between karst water resources and developed karst areas in each region, can also help in the optimal management of water resources. Classification using different features for specific classes as one of the advantages of the object-oriented method has made this method an efficient and reliable method. Due to the specific tectonic and lithological conditions that have been mentioned, the conditions for the development of dissolution and morphological processes exist in the surface and depth, and according to the researches, there are karst processes with climatic conditions at altitudes above 2500 meters. This method is suitable for detecting sinkholes and also makes it possible to identify areas close to aquifers in large karst massifs. Such studies can identify areas with potential for water resources in these formations to enable subsurface exploration in areas that are inaccessible or impassable.