Determining Meaningful Electrical Conductivity Values from In Situ Substrate Measurements

Electrical conductivity (EC) measurements in both soils and soilless substrates have been used to make fertilizer decisions and to manage salt buildup. Pore water electrical conductivity (EC_p) cannot be measured directly in substrates, and is often modeled from in-situ measurements of bulk EC (EC_b), substrate water content (sometimes represented as a single dielectric permittivity (ε) value), and a variety of substrate-specific coefficients. Our group analyzed data from three different soilless substrate types as well as stonewool to better understand the limitations and the robustness of two models to predict EC_p from EC_b and dielectric permittivity.  In addition, we collected εvalues measured at different frequencies in stonewool to determine if additional parameters could improve model accuracy. Our data suggest that existing models for EC_p are most accurate (10%) at very high volumetric water contents with as much as 100% error at lower volumetric water contents. Our multifrequency data in stonewool allowed us to develop a new approach, where we were able to get an EC_p prediction independent of a bulk EC measurement. The success of the mutifrequency approach in stonewool is promising for a new, more accurate EC_p model for both soils and soilless substrates.