Premium
Measurement of volumetric water content by TDR in saline soils
Author(s) -
WYSEURE G.C.L.,
MOJID M.A.,
MALIK M.A.
Publication year - 1997
Publication title -
european journal of soil science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.244
H-Index - 111
eISSN - 1365-2389
pISSN - 1351-0754
DOI - 10.1111/j.1365-2389.1997.tb00555.x
Subject(s) - reflectometry , soil water , water content , dielectric , electrical resistivity and conductivity , soil science , soil salinity , conductivity , calibration , materials science , electrical impedance , saline water , salinity , time domain , environmental science , chemistry , geotechnical engineering , mathematics , geology , physics , optoelectronics , statistics , quantum mechanics , oceanography , computer science , computer vision
Summary Time‐domain reflectometry (TDR) evaluates the bulk dielectric constant, K , of the soil by measuring the travel time of an electromagnetic pulse through a sensor, and through it estimates the volumetric water content. We show that for saline soils the effects of conductivity and frequency on the travel time cannot be neglected and that, as a result, TDR systematically overestimates the water content in saline soils. Simultaneously the bulk electrical conductivity of soils can be estimated by TDR. The equivalent impedance after multiple reflections is related to the bulk electrical conductivity, σ This relation differs from sensor to sensor and requires calibration for each individual sensor. A method is proposed for correcting the volumetric water content in saline soils. First, the bulk electrical conductivity, o , is estimated from the equivalent impedance at a specific equivalent distance of cable, several times the actual length of the sensor. The zero‐salinity dielectric constant, K O , of this soil is obtained by correcting the apparent K as a function of the measured bulk electrical conductivity. The volumetric water content is estimated from K o . The correction of K is a function of the equivalent frequency of the electromagnetic pulse. The imaginary part of the dielectric constant is primarily due to ohmic losses. The model, which calculates the velocity of propagation of the electromagnetic pulse and which takes into consideration the imaginary part, performs reasonably well. An empirical approach based on calibration gave slightly better results.