Soil salinity mapping using Landsat 8 OLI data and regression modeling in the Great Hungarian Plain

Abstract Salt's deposition in the subsoil is known as salinization. It caused by natural processes such mineral weathering or human-made activities irrigation with saline water. This environmental issue has grown more critical and frequently occurring Hungarian Great Plain, adversely influencing agricultural productivity. study aims to predict soil salinity located east of Hungary, using Landsat 8 OLI data combined four state-of-the-art regression models, i.e., Multiple Linear Regression, Partial Least Squares Ridge Feedforward Artificial Neural Network. For this purpose, seventy-six samples were collected during a field survey conducted Research Institute for Soil Sciences Agricultural Chemistry between 15 September October, 2016. We used min–max accuracy, root-mean-square error (RMSE), mean squared (MSE) evaluate compare models' performance. The results showed that ridge model performed best terms prediction (MSE training = 0.006, MSE test 0.0007, RMSE 0.081), accuracy equal 0.75. Hence, application modeling on spectral indices, principal component analysis, land surface temperature derived from multispectral an efficient method assessment at local scales. resulting map can provide overview levels efficiency management strategies irrigated areas. An increase sampling density will be recommended validate approach regional scale.