There are several ways to express, but most commonly used is weight of the water divided by weight of dry soil.  So if the initial sample weighs 40 g, and the weight of water is 20 g and dry soil is also 20 g, the value is 1 or 100%.  organic material in some soils can hold 4 times their weight in water, so theoretically, the value could be even larger than 100%.  This is called something like the dry weight approach.  There are also wet weight methods and volumetric methods, so your sensor should explain what it referring to and how its calibration can be checked, and at what frequency is suggested to maintain research or study level desired accuracy and results.  You should report how the sensor is expressing soil moisture in the methods section, and if important, could be a footnote in tables if it reports soil water in different units, such as by volume.

Sorry, I forgot to add, salt water is heavier than fresh water. I don't know how this may factor in to your analysis, and maybe a good question for tech support from the company. ,When you evaporate water from a soil sample, I think the salt would be retained as part of the soil weight.  You may try to take a salt water sample, weigh it, evaporate the water and weigh the residual salt to get an idea of its influence.

I have not used this particular brand of sensors; however, there may be an issue with the calibration curve. Have you calibrated your sensors to this particular soil?  The calibration curves  provided by the manufacturer is good for mineral soils having an specific range of electrical conductivity.  In this case between 0.1 - 10 dS/m.  For soils outside this range, you may need to calibrate the sensors before use.  We have done this for other type of sensors such as CS615, CS 616, but not for these sensors you mentioned.

I do not agree soil volumetric content up to 100% until whole soil gets saturated to 100%. It is rare or impossible. Kindly calibrate the sensors and try again.

Dear All,

Thank you very much for your reply and suggestions. Yes, before the installation of the sensors, we did sensor calibration by soil column experiments using soil samples from the research area, yet  one thing we did not take into account that we should calibrate each sensor for the corresponding soil layer, because soil profiles and groundwater depth are very different in these soil monitoring sites.

Best regards,

Dear Abliz. One way to check if the soil moisture values obtained from the sensor, actually correspond to ground condition in which you are evaluating is to make the comparison of the values obtained from the sensor that you mentioned, with the values obtained by the method standard greenhouse. Place the soil sample in an aluminum can with known weight, closes well, so there is no water vapor output. Thereafter, the assembly is weighed to yield wet weight. it takes to dry the soil samples in an oven at 105-110 ° C for 24 hours. After drying the samples should be placed in desiccator to cool without moisture absorption and subsequently weighed. And applied the formula: Ug= (Mu-Ms)/ Msx100, where: Ug is the gravimetric water content in soil (% by weight), Mu is the wet weight of sample (g), and Ms is the dry weight of sample (g).

   Dear Abdulla Abliz,

   The values you might be concerned are fine (a soil retaining more water than its mass). Take a look at figure 2D of the attached paper. It might help you getting more comfortable with that.

   Marcelo Barbosa is also right. You have to calibrate your sensor with a water retention curve. If you do not rely on the sensor, and if your protocol allows, measure it gravimetrically. It will be much more reliable.

   Good luck,

   Gustavo

Dear Abdullah,

Perhaps the attached paper might help you figure this out. I also saw on the Decagon website that salinity can vary your readings by up to 10%, so that might be a contributing factor.

Regards,

Steve

Dear Abdullah,

Perhaps the attached paper might help you figure this out. I also saw on the Decagon website that salinity can vary your readings by up to 10%, so that might be a contributing factor.

Regards,

Steve