دانشگاه کاشان مهندسی اکوسیستم بیابان 2538-6336 7 شماره 1 انگلیسی 2018 08 23 Comparison of artificial neural network and multiple linear regressions efficiency for predicting soil salinity in Yazd -Ardakan plain, central Iran Comparison of artificial neural network and multiple linear regressions efficiency for predicting soil salinity in Yazd -Ardakan plain, central Iran 20 11 114049 FA فاطمه روستایی Combat to desertification, Natural Resource Faculty, Ardakan University, Yazd, Iran شمس اله اتوبی Department of soil science, College of Agriculture, Isfahan University of Technology, Isfahan, Iran مجتبی نوروزی Department of Soil Science, College of Agriculture, Shahid Chamran University of Ahvaz, Khuzestan, Iran Journal Article 2017 12 22 The study was conducted to evaluate the efficacy of artificial neural network (ANN) and multivariate regression (MLR) analysis to predict spatial variability of soil salinity in central Iran, using remotely sensed data. The analysis was based on data acquired from EOS AMI remote sensing satellite. The two methods was used to study linear and non-linear relationship between soil reflectance and soil salinity. In MLR analysis, stepwise method and neural network were applied using sensitivity coefficient by arranging inputs through the backward propagation, and then modeling was done. The R² and RMSE were 0.23 and 0.33 for MLR, and 0.79 and 0.11 for ANN, respectively. Digital values of VNIR1 and NDVI48 were identified as the most important factors in MLR, whereas Sum19 and SWIR6 were recognized as the most important data to predict soil salinity using ANN. The results indicated that ANN model is used to detect non- linear relationship between soil salinity and ASTER data at the study area. The study was conducted to evaluate the efficacy of artificial neural network (ANN) and multivariate regression (MLR) analysis to predict spatial variability of soil salinity in central Iran, using remotely sensed data. The analysis was based on data acquired from EOS AMI remote sensing satellite. The two methods was used to study linear and non-linear relationship between soil reflectance and soil salinity. In MLR analysis, stepwise method and neural network were applied using sensitivity coefficient by arranging inputs through the backward propagation, and then modeling was done. The R² and RMSE were 0.23 and 0.33 for MLR, and 0.79 and 0.11 for ANN, respectively. Digital values of VNIR1 and NDVI48 were identified as the most important factors in MLR, whereas Sum19 and SWIR6 were recognized as the most important data to predict soil salinity using ANN. The results indicated that ANN model is used to detect non- linear relationship between soil salinity and ASTER data at the study area. https://deej.kashanu.ac.ir/article_114049_0806c5dc2c16fbdcdc163048778c96c9.pdf