Spatial Modeling of Groundwater Resources Potential in Telvar Watershed using Support Vector Machine and Random Forest Models

Message:
Article Type:
Research/Original Article (دارای رتبه معتبر)
Abstract:

In this research, it is tried to identify the potential status of groundwater resources in different parts of the Telvar watershed using two machine learning models including support vector machine and random forest models. Initially, information about the wells in the region was received from the Regional Water Company of Kurdistan. The wells in the area were randomly divided into two groups of training (including 70% of data) and validation (including 30% of data). Elevation, slope, slope direction, lithology, pedology, surface curvature, land use, topographic moisture index and distance from the river were selected as predictor variables and their map was prepared in the GIS environment. The data of the training group along with the maps related to the predictor variables were entered into the support vector machine model and the random forest model. Based on the data of the training group, the parameters of the model were calibrated and adjusted and the potential of groundwater resources was predicted. The prediction accuracy of the models was determined using the statistical method of performance characteristic curve in two stages of training and validation. The results showed that accuracy of the random forest model (98.4%) was more than the support vector machine model (98.1%).

Language:
Persian
Published:
Whatershed Management Research, Volume:35 Issue: 134, 2022
Pages:
73 to 87
https://www.magiran.com/p2420797  
سامانه نویسندگان
  • Corresponding Author (2)
    Mohammad Sedigh Ghorbani
    Assistant Professor Physical Geography(Geomorphology), Department of Social Science, Payame Noor University, Payame Noor University, Tehran, Iran
    Ghorbani، Mohammad Sedigh
اطلاعات نویسنده(گان) توسط ایشان ثبت و تکمیل شده‌است. برای مشاهده مشخصات و فهرست همه مطالب، صفحه رزومه را ببینید.
مقالات دیگری از این نویسنده (گان)