|Title||Field tests of dielectric sensors in a facility for studying salt tolerance of crops|
|Author(s)||Straten, Gerrit Van; Vos, Arjen De; Vlaming, Rik; Oosterbaan, Roland|
|Source||International Agricultural Engineering Journal 25 (2016)2. - ISSN 0858-2114 - p. 102 - 113.|
|Department(s)||Biobased Chemistry and Technology|
|Publication type||Refereed Article in a scientific journal|
|Keyword(s)||Calibration - Electrical conductivity - Irrigation - Salinity - Salt tolerance - Soil sensor - Volumetric water content|
Seven salinity levels ranging from 1.7 to 35 dS/m are applied to groups of eight fields each in a field facility for testing the salt tolerance of crops. Each of the 56 test fields is equipped with one or two dielectric sensors for soil volumetric water content (VWC) and bulk electric conductivity (ECb). Several models for calibrating the sensors in the laboratory were tested and parameterized. Overall, the root mean square error was in the range of 0.57-0.59 dS/m in terms of soil bulk EC. The models differed in their robustness against inversion to obtain pore water EC from measured bulk EC. The laboratory calibration formula overestimates the pore water EC at low EC (5 dS/m), and underestimates it at high EC (25 dS/m). In practice, calculated sensor pore water EC's in fields with the same salinity treatment differ among each other, showing the limitations of laboratory calibrations. However, in fields where pore water samples are available, a direct proportionality between pore water EC and sensor bulk EC suffices without correction for VWC in this well irrigated case. Moreover there is a good correlation between the low frequent EC time series of suction cup samples and the high frequent sensor readings. When used with care, sensors can give valuable information about the dynamics of soil conditions during crop salinity tolerance tests.