XML Persian Abstract Print


Abstract:  
Groundwater resources are always one of the most important and most reliable sources of water in arid and semi-arid regions and the extraction of water from these sources is of particular importance to other climates. Therefore, the purpose of this study is to provide a groundwater potential map in Jahrom city, Fars province using Shannon entropy methods and random forest log algorithm and compare their accuracy. According to the geology, hydrological, structural and topographic data required for research, maps were prepared in ArcGIS 10.3 and SAGA format with the Raster format. Then, using Shannon entropy model and random forest model, the weight of each of the effective factors in R software was calculated and finally the groundwater potential map was prepared. After designing the spring potential maps using these methods, ROCs were used to evaluate the results. The surface area below the curve (AUC) obtained from the relative yield recognition curve shows a precision of 90.83 and 76.20% respectively for Shannon entropy models and random forest. The results show the high accuracy of Shannon entropy model compared to the random forest model.
     
Type of Study: Research | Subject: Sciences and Watershed Engineering
Received: 2017/12/9 | Accepted: 2019/07/6

Send email to the article author


© 2019 All Rights Reserved | Desert Ecosystem Engineering Journal

Designed & Developed by : Yektaweb