Predicting soil EC_e based on values of EC_1:2.5 as an indicator of soil salinity at Magozi Irrigation Scheme, Iringa, Tanzania

Abstract

Soil salinity is one of the limitations to sustainable production of rice and other crops in many irrigation schemes in Tanzania. Soil salinity can be assessed from electrical conductivity (EC) measurements. Most soil laboratories in Tanzania appraise soil salinity from measurements of electrical conductivity of 1:2.5 soil:water suspensions (EC_1:2.5) by virtue of their simplicity. However, the influence of soil salinity on plant growth is mainly based on electrical conductivity of saturated paste extract (EC_e), so it is necessary to convert EC_1:2.5 to EC_e in order to assess plant response to salinity. This study was conducted at Magozi Irrigation Scheme in Iringa Region, Tanzania to establish regression model for predicting EC_e from EC_1:2.5 values. A total of 60 soil samples (45 samples for model training and 15 samples for model validation) were collected and analyzed for soil EC_1:2.5, EC_e and soil texture. Results showed that EC_1:2.5 ranged from 0.1 to 4.2 dS m^-1 with a mean value of 0.71 dS m^-1. EC_e obtained ranged from 0.3 (non-saline) to 12 dS m^-1 (very saline) with a mean of 2.4 dS m^-1 (slightly saline). In order of dominance, soil textural classes were sandy clay loam, clay, sandy clay, sandy loam and clay loam. Strong linear relationships between EC_e and EC_1:2.5 were observed in the developed linear regression equations. After validation, the study selected equation EC_e = 3.4954*EC_1:2.5 with R² of 0.956 for combined soil textures to be used for prediction of EC_e from EC_1:2.5 at Magozi Irrigation Scheme. This model can be tested for its applicability to other similar soils in Tanzania in further studies.