To establish sustainable agricultural practices, monitoring soil water and solute distributions in actual field conditions is important. Because the estimation of electrical conductivity (EC) using time domain reflectometry (TDR) is considered to be affected by Andisols because of their unique dielectric properties, in this study, we investigate the potential for the continuous monitoring of EC and nitrate concentrations using TDR in an Andisol field. The study was performed from December 2007 to August 2008. To estimate soil solution EC (EC_w), the relationship among EC_w, apparent EC (EC_a) and volumetric water content (θ) for an Andisol was investigated, and three models (Rhoades model, Extended Rhoades model and Hilhorst model) were applied to describe this relationship. The Rhoades model was found to be reasonably accurate in describing the EC_w-EC_a-θ relationship, while the Hilhorst model showed reasonable agreement with the experimental data. In the field experiment, EC_w values estimated using TDR with the Rhoades model correlated with those obtained from solution samples and a linear regression between EC_w and soil nitrate concentrations was obtained from the field experimental data. The use of this regression with EC_w estimated from TDR measurements, together with the Rhoades model, is considered a useful tool for the continuous monitoring of soil nitrate concentrations in Andisol fields under transient conditions.