RF communication through salt water

You can do it by calculation. The conductivity of the salt water will allow you to derive the 'skin depth' for the frequency of transmission. Each skin depth is an attenuation of 8.6dB (it's 1/e, by definition). You should be able to find figures for the effective transmit signal and the minimum receive signal for your module and the difference, your link budget, can then be used to determine the range.

Quick and dirty version: Wikipedia gives a skin depth of 25cm at 1MHz for seawater; that would equate to no more than a centimetre or two at 433MHz; your link budget is maybe 8 x 8.6dB; so you might manage a range of 10cm-20cm.

You can do it by calculation. The conductivity of the salt water will allow you to derive the 'skin depth' for the frequency of transmission. Each skin depth is an attenuation of 8.6dB (it's 1/e, by definition). You should be able to find figures for the effective transmit signal and the minimum receive signal for your module and the difference, your link budget, can then be used to determine the range.

Quick and dirty version: Wikipedia gives a skin depth of 25cm at 1MHz for seawater; that would equate to no more than a centimetre or two at 433MHz; your link budget is maybe 8 x 8.6dB; so you might manage a range of 10cm-20cm.

How does that compare to fresh water?

C

If the water were pure there would be no conduction, so shielding by skin effect goes away, but the water is a dielectric so it would still absorb some of the radiation.

Overall, the water molecule is electrically neutral, but the way the atoms and electrons arrange themselves it looks from the outside as though there is an electric dipole there - the 'centre of gravity' of the positive charges looks to be in a different place to that of the negative ones. A passing electromagnetic wave will give up energy to physically swing the molecule back and forth [effectively, it's heating it up].

Our measure of that, as engineers, is relative permittivity [sometimes called the dielectric constant]. It will vary with the temperature and pressure of the water and with the frequency of the wave, so it's not easy to give a simple answer.

Wikipedia gives the static permittivity of water as 80.1 at 20 degrees C. That's for very low frequencies (hence the 'static') and presumably the pressure you'd get with a beaker of water in a laboratory. I don't know how that changes at 433MHz.

That's much higher than I would have guessed [don't know why - I've seen a microwave cooking food, after all], so it's going to make the situation for imomo0 even worse. Against that, I was probably far too pessimistic about the link budget, so it won't necessarily make too much difference to my rough, back-of-the-envelope calculation.

Its fascinating!  I actually saw a demonstration once with conductivity between tap water and distilled water, and if i remember right it was the distilled that would not pass a charge ... I think the explanation was that its the chemicals in the water that pass the current!  but I'm stating what everyone else probably already knows