Estimating Electrical Conductivity for Soil Salinity Monitoring Using Various Soil-Water Ratios Depending on Soil Texture

Abstract

ABSTRACTThe electrical conductivity of a saturated paste extract of soil (ECe) is a standard measurement of soil salinity, which may adversely effect the environment and plants. This study aimed to develop a precise linear regression relationship between ECe and electrical conductivity (EC) of different (1:5, 1:2.5, 1:1) soil-water ratios based on soil texture (represented by an index Txw) for the ease and rapid monitoring of soil salinity in an area. Surface (0–20 cm) soil samples (n = 150) from the coastal zone of Bangladesh were analyzed for particle size distribution, and EC by various methods. Entire soil samples were equally divided into two textural groups (finer and coarser) based on Txw values. Necessary statistical treatments were performed to satisfy the conditions of linear regression analysis. Significant correlations (r = −0.44** to −0.56**) were found between Txw and the EC values of various methods within the whole dataset. Derived general (not Txw-assisted, i.e., all data) and specific (Txw-assisted groups) equations followed the model ECe =. Almost all specific equations showed improved coefficient of determinations (mean of all r2 = 0.86; all p < .0001) compared to general ones (mean of all r2 = 0.81; all p < .0001). While validation, corresponding errors (RMSE and/or ME) were found less in the specific equations than the general ones in predicting ECe. Therefore, soil texture (Txw)-based derived equations can preferably be used to predict ECe using other methods for the monitoring of soil salinity in an area.